Dell 3 User Manual
Dell™ PowerEdge Expandable RAID Controller 3/QC, 3/DC, 3/DCL and  
3/SC  
PERC 3 User’s Guide  
w w w. d e l l . c o m | s u p p o r t . d e l l . c o m  
Safety Instructions  
CAUTION: Safety Instructions  
Use the following safety guidelines to help ensure your own personal safety  
and to help protect your computer and working environment from potential  
damage.  
General  
•
Do not attempt to service the computer yourself unless you are a  
trained service technician. Always follow installation instructions  
closely.  
•
To help prevent electric shock, plug the computer and device power  
cables into properly grounded electrical outlets. These cables are  
equipped with 3-prong plugs to help ensure proper grounding. Do not  
use adapter plugs or remove the grounding prong from a cable. If you  
must use an extension cable, use a 3-wire cable with properly grounded  
plugs.  
•
•
To help avoid the potential hazard of electric shock, do not use your  
computer during an electrical storm.  
To help avoid the potential hazard of electric shock, do not connect or  
disconnect any cables or perform maintenance or reconfiguration of  
this product during an electrical storm.  
•
•
If your computer includes a modem, the cable used with the modem  
should be manufactured with a minimum wire size of 26 American  
wire gauge (AWG) and an FCC-compliant RJ-11 modular plug.  
Before you clean your computer, disconnect the computer from the  
electrical outlet. Clean your computer with a soft cloth dampened  
with water. Do not use liquid or aerosol cleaners, which may contain  
flammable substances.  
•
•
To help avoid possible damage to the system board, wait 5 seconds  
after turning off the computer before disconnecting a device from the  
computer.  
To avoid shorting out your computer when disconnecting a network  
cable, first unplug the cable from the network adapter on the back of  
your computer, and then from the network jack. When reconnecting a  
network cable to your computer, first plug the cable into the network  
jack, and then into the network adapter.  
1
•
To help protect your computer from sudden, transient increases and  
decreases in electrical power, use a surge suppressor, line conditioner,  
or uninterruptible power supply (UPS).  
•
•
•
Ensure that nothing rests on your computer’s cables and that the  
cables are not located where they can be stepped on or tripped over.  
Do not push any objects into the openings of your computer. Doing so  
can cause fire or electric shock by shorting out interior components.  
Keep your computer away from radiators and heat sources. Also, do not  
block cooling vents. Avoid placing loose papers underneath your  
computer; do not place your computer in a closed-in wall unit or on a  
bed, sofa, or rug.  
When Using Your Computer  
As you use your computer, observe the following safe-handling guidelines.  
CAUTION: Do not operate your computer with any cover(s)  
(including computer covers, bezels, filler brackets, front-panel  
inserts, and so on) removed.  
Your computer is equipped with one of the following:  
•
A fixed-voltage power supply — Computers with a fixed-voltage power  
supply do not have a voltage selection switch on the back panel and  
operate at only one voltage (see the regulatory label on the outside of  
the computer for its operating voltage).  
•
•
An auto-sensing voltage circuit — Computers with an auto-sensing  
voltage circuit do not have a voltage selection switch on the back panel  
and automatically detect the correct operating voltage.  
A manual voltage selection switch — Computers with a voltage  
selection switch on the back panel must be manually set to operate at  
the correct operating voltage.  
To help avoid damaging a computer with a manual voltage selection switch,  
ensure that the voltage selection switch is set to match the AC power  
available at your location:  
•
115 V/60 Hz in most of North and South America and some Far  
Eastern countries such as South Korea and Taiwan  
2
•
•
100 V/50 Hz in eastern Japan and 100 V/60 Hz in western Japan  
NOTE: The voltage  
selection switch must be  
set to the 115-V position  
even though the AC power  
available in Japan is  
100 V.  
230 V/50 Hz in some regions in the Caribbean and South America and  
most of Europe, the Middle East, and the Far East  
Also, ensure that your monitor and attached devices are electrically rated to  
operate with the AC power available in your location.  
•
Before working inside the computer, unplug the computer to help  
prevent electric shock or system board damage. Certain system board  
components continue to receive power any time the computer is  
connected to AC power.  
When Working Inside Your Computer  
Before you open the computer cover, perform the following steps in the  
sequence indicated.  
CAUTION: Do not attempt to service the computer yourself,  
except as explained in your online Dell™ documentation or in  
instructions otherwise provided to you by Dell. Always follow  
installation and service instructions closely.  
NOTICE: To help avoid possible damage to the system board, wait 5 seconds  
after turning off the computer before removing a component from the system  
board or disconnecting a device from the computer.  
1
Perform an orderly computer shutdown using the operating system  
menu.  
2
3
Turn off your computer and any devices connected to the computer.  
Ground yourself by touching an unpainted metal surface on the  
chassis, such as the metal around the card-slot openings at the back of  
the computer, before touching anything inside your computer.  
While you work, periodically touch an unpainted metal surface on the  
computer chassis to dissipate any static electricity that might harm  
internal components.  
4
Disconnect your computer and devices, including the monitor, from  
their electrical outlets. Also, disconnect any telephone or  
telecommunication lines from the computer.  
Doing so reduces the potential for personal injury or shock.  
In addition, take note of these safety guidelines when appropriate:  
3
•
•
When you disconnect a cable, pull on its connector or on its strain-  
relief loop, not on the cable itself. Some cables have a connector with  
locking tabs; if you are disconnecting this type of cable, press in on the  
locking tabs before disconnecting the cable. As you pull connectors  
apart, keep them evenly aligned to avoid bending any connector pins.  
Also, before you connect a cable, ensure that both connectors are  
correctly oriented and aligned.  
Handle components and cards with care. Do not touch the  
components or contacts on a card. Hold a card by its edges or by its  
metal mounting bracket. Hold a component such as a microprocessor  
chip by its edges, not by its pins.  
CAUTION: There is a danger of a new battery exploding if it is  
incorrectly installed. Replace the battery only with the same or  
equivalent type recommended by the manufacturer. Do not  
dispose of the battery along with household waste. Contact your  
local waste disposal agency for the address of the nearest battery  
deposit site.  
Protecting Against Electrostatic Discharge  
Static electricity can harm delicate components inside your computer. To  
prevent static damage, discharge static electricity from your body before you  
touch any of your computer’s electronic components, such as the  
microprocessor. You can do so by touching an unpainted metal surface on  
the computer chassis.  
As you continue to work inside the computer, periodically touch an  
unpainted metal surface to remove any static charge your body may have  
accumulated.  
You can also take the following steps to prevent damage from electrostatic  
discharge (ESD):  
•
Do not remove components from their antistatic packing material  
until you are ready to install the component in your computer. Just  
before unwrapping the antistatic packaging, discharge static electricity  
from your body.  
•
•
When transporting an electrostatic sensitive component, first place it  
in an antistatic container or packaging.  
Handle all electrostatic sensitive components in a static-safe area. If  
possible, use antistatic floor pads and workbench pads.  
4
Ergonomic Computing Habits  
CAUTION: Improper or prolonged keyboard use may result in  
injury.  
CAUTION: Viewing the monitor screen for extended periods of  
time may result in eye strain.  
Battery Disposal  
Do not dispose of the battery along with household waste. Contact your  
local waste disposal agency for the address of the nearest battery deposit  
site.  
5
6
Contents  
Safety Instructions .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
1 Overview  
PERC 3 Overview .  
.
24  
PERC 3 Features  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
24  
SCSI Channels  
25  
Non-volatile Random Access Memory (NVRAM) and Flash  
Read-only Memory (ROM)  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
25  
25  
26  
26  
26  
27  
SCSI Connectors  
.
.
.
.
.
.
.
.
.
.
Single-Ended and LVD SCSI Buses .  
Maximum Cable Length for SCSI Standards  
SCSI Bus Widths and Maximum Throughput .  
Operating System Support  
.
.
.
.
.
.
.
.
.
.
2 Introduction to RAID  
RAID Definition .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
30  
30  
PERC 3 Host-Based RAID Solution  
Components and Features  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
31  
31  
31  
31  
32  
32  
34  
35  
36  
Physical Array  
Logical Drive  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Fault Tolerance .  
Consistency Check .  
Disk Striping  
.
.
.
.
.
.
.
Disk Mirroring  
Disk Spanning  
Spanning for RAID 10 or RAID 50  
Contents  
7
Parity  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
36  
37  
38  
39  
39  
40  
40  
Hot Spares .  
Disk Rebuilds  
Hot Swap  
.
.
SCSI Drive States .  
Logical Drive States .  
Enclosure Management  
3 RAID Levels  
Overview .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
42  
42  
Selecting a RAID Level  
RAID 0 .  
RAID 1 .  
RAID 5 .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
43  
45  
46  
RAID 10  
RAID 50  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
48  
50  
4 PERC 3/SC Features  
Hardware Requirements .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
54  
Configuration Features  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
54  
55  
55  
SMART Technology  
Configuration on Disk Configuration .  
Hardware Architecture Features  
Array Performance Features  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
56  
56  
57  
57  
58  
.
.
PERC 3/SC Fault Tolerance Features  
Software Utilities  
Operating System Software Drivers .  
.
.
.
.
.
.
.
.
.
.
8
Contents  
PERC 3/SC Specifications .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
58  
59  
59  
60  
60  
60  
60  
60  
60  
61  
61  
PCI Bridge/CPU .  
Cache Memory .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
PERC 3/SC BIOS  
Onboard Speaker  
Serial Port .  
SCSI Bus  
.
.
.
.
.
.
.
SCSI Connectors .  
SCSI Termination  
SCSI Firmware  
.
RAID Management  
5 PERC 3/DC and PERC 3/DCL Features  
Hardware Requirements .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
64  
Configuration Features  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
64  
65  
65  
SMART Technology  
Configuration on Disk  
Hardware Architecture Features  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
66  
67  
67  
68  
68  
Array Performance Features  
.
.
Fault Tolerance Features .  
Software Utilities .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Operating System Software Drivers .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
PERC 3/DC and PERC 3/DCL Specifications  
.
.
.
.
.
.
.
.
.
.
.
68  
69  
70  
70  
70  
70  
70  
71  
71  
PCI Bridge/CPU .  
Cache Memory .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
PERC 3/DC and PERC 3/DCL BIOS .  
Onboard Speaker  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Serial Port .  
SCSI Bus  
.
.
.
.
.
.
.
SCSI Connectors .  
SCSI Termination  
9
Contents  
SCSI Firmware  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
71  
72  
RAID Management  
6 PERC 3/QC Features  
Hardware Requirements .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
74  
Configuration Features  
.
.
74  
75  
75  
SMART Technology  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Configuration on Disk  
Hardware Architecture Features  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
76  
76  
77  
77  
78  
Array Performance Features  
.
.
.
.
.
.
Fault Tolerance Features .  
Software Utilities .  
.
.
.
.
.
.
.
Operating System Software Drivers .  
PERC 3/QC Specifications  
.
.
.
.
.
78  
79  
79  
80  
80  
80  
80  
80  
80  
81  
81  
PCI Bridge/CPU .  
Cache Memory .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
PERC 3/QC BIOS  
Onboard Speaker  
Serial Port .  
SCSI Bus  
.
.
.
.
.
.
.
SCSI Connectors .  
SCSI Termination  
SCSI Firmware  
.
RAID Management  
7 Configuring PERC 3  
Configuring SCSI Physical Drives .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
84  
84  
Current Configuration .  
.
.
.
.
.
.
.
.
.
Logical Drive Configuration .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
87  
10  
Contents  
Physical Device Layout  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
89  
Configuring Arrays  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
92  
93  
93  
93  
Arranging Arrays  
Creating Hot Spares .  
Creating Logical Drives  
Configuration Strategies  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
93  
94  
94  
95  
Maximizing Capacity  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Maximizing Drive Availability  
Maximizing Drive Performance  
Assigning RAID Levels  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
95  
96  
Configuring Logical Drives  
Optimizing Data Storage  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
96  
96  
97  
Data Access Requirements .  
Array Functions .  
.
.
.
.
.
Planning the Array Configuration .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
97  
Using the Array Configuration Planner .  
98  
Random Array Deletion  
Overview .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
99  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
99  
Configuration Module  
. 100  
8 PERC 3/SC Hardware Installation  
Requirements  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 102  
. 102  
Optional Equipment  
PERC 3/SC Card Layout .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 103  
Installation Steps  
.
.
.
.
.
.
. 103  
. 104  
. 104  
. 105  
. 108  
Step 1—Unpack .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Step 2—Power Down  
Step 3—Set Jumpers  
Step 4—Set SCSI Termination .  
11  
Contents  
Step 5—Install PERC 3/SC  
Step 6—Connect SCSI Cables .  
Step 7—Set Target IDs  
Step 8—Power On Host System  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 111  
. 113  
. 114  
. 115  
.
.
.
.
Step 9—Run PERC 3 BIOS Configuration Utility or WebBIOS  
Utility .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 116  
. 116  
Step 10—Install Operating System Software Drivers .  
.
.
.
.
9 PERC 3/DC or PERC 3/DCL Hardware Installation  
Requirements  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 118  
. 118  
Optional Equipment  
PERC 3/DC Card Layout .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 119  
PERC 3/DCL Card Layout .  
Installation Steps  
. 120  
.
.
.
.
.
. 120  
. 121  
. 121  
. 121  
. 125  
. 126  
. 128  
. 128  
. 130  
Step 1—Unpack the PERC 3/DC or PERC 3/DCL  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Step 2—Power Down  
Step 3—Set Jumpers  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Step 4—Set SCSI Termination .  
Step 5—Install the PERC 3/DC or PERC 3/DCL Controller  
Step 6—Select and Set Target IDs for SCSI Devices .  
.
.
.
.
.
.
.
.
.
Step 7—Connect SCSI Cables .  
Step 8—Power On Host System  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Step 9—Run the PERC 3 BIOS Configuration Utility or WebBIOS  
Configuration Utility  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 130  
Step 10—Install Operating System Software Driver  
.
.
.
.
.
. 131  
Replacing a PERC 3/DC Containing a BC Chip with a PERC 3/DC  
Containing a BE Chip . 131  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Using Driver 5.22.1 or 5.22.2 and Firmware 161J or 161N . . 131  
Using Driver 5.30 and Firmware 1.70 .  
.
.
.
.
.
.
.
.
.
.
.
. 131  
10 PERC 3/QC Hardware Installation  
Requirements  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 134  
12  
Contents  
Optional Equipment  
PERC 3/QC Card Layout .  
Installation Steps  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 134  
. 135  
.
.
.
.
. 136  
. 136  
. 137  
. 137  
. 140  
. 141  
. 143  
. 145  
. 146  
. 146  
Step 1—Unpack the PERC 3/QC Controller  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Step 2—Power Down  
Step 3—Set Jumpers  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Step 4—Set SCSI Termination .  
SCSI Termination  
.
.
.
.
.
.
.
Step 5—Install PERC 3/QC  
.
.
Step 6—Connect SCSI Cables .  
Step 7—Set Target IDs  
Step 8—Power on Host System  
.
.
.
.
Step 9—Run the PERC 3 BIOS Configuration Utility or WebBIOS  
Configuration Utility .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 147  
Step 10—Install Operating System Driver  
.
.
.
.
.
.
.
.
.
.
. 147  
11 PERC 3 BIOS Configuration Utility  
Configuration On Disk .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 150  
Starting the PERC 3 BIOS Configuration Utility .  
. 150  
BIOS Configuration Utility Menu Options  
.
.
.
.
. 151  
. 152  
. 153  
. 153  
. 158  
. 158  
. 159  
Configure Menu  
Initialize Menu  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Objects Menu  
Format Menu  
Rebuild Menu  
.
.
.
Check Consistency Menu .  
Configuring Arrays and Logical Drives  
Choosing the Configuration Method  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 159  
.
. 160  
. 160  
. 160  
. 161  
Easy Configuration .  
New Configuration .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
View/Add Configuration  
13  
Contents  
PERC 3/QC and PERC 3/DC Default Settings  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 161  
. 161  
. 162  
PERC 3/SC and PERC 3/DCL Default Settings .  
Reserved Disk Space during Configuration .  
.
.
.
.
.
.
.
.
.
.
Designating Drives as Hot Spares .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 162  
. 163  
. 167  
. 173  
Using Easy Configuration .  
Using New Configuration  
.
.
.
.
.
.
.
.
.
.
.
Using View/Add Configuration  
Initializing Logical Drives .  
Batch Initialization  
Individual Initialization  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 178  
. 178  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 179  
Deleting Logical Drives (Random Array Deletion) .  
.
.
.
.
.
.
.
. 179  
Formatting Physical Drives  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 180  
. 180  
. 180  
Media Errors  
.
.
.
.
.
.
.
.
Formatting Drives  
Rebuilding Failed Disk Drives .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 182  
. 182  
. 182  
Manual Rebuild – Rebuilding an Individual Drive .  
Manual Rebuild – Batch Mode .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Using a Pre-loaded SCSI Drive “As-is” .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 183  
. 184  
. 184  
Exiting the PERC 3 BIOS Configuration Utility  
Setting Hardware Termination  
.
.
.
.
.
.
.
.
.
Clustering  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 185  
. 185  
. 185  
Cluster Mode  
Enabling and Disabling the Cluster Mode  
12 Dell Manager  
Starting Dell Manager .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. 188  
. 189  
Management Menu Options  
Using Dell Manager in Red Hat Linux GUI Mode  
.
.
.
.
.
.
. 189  
14  
Contents  
Dell Manager Menu Options  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
190  
190  
190  
194  
195  
195  
195  
Initialize Menu  
Objects Menu .  
Format Menu .  
Rebuild Menu .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Check Consistency Menu  
Reconstruct Menu .  
.
.
.
Designating Drives as Hot Spares  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
195  
196  
Parameters  
.
.
.
.
.
.
.
.
.
.
.
.
Initializing Logical Drives  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
197  
197  
198  
Batch Initialization .  
.
Individual Initialization .  
Deleting Logical Drives (Random Array Deletion)  
.
.
.
.
.
.
.
.
198  
Formatting Physical Drives .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
199  
199  
200  
Media Errors .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Formatting Drives .  
Rebuilding Failed Disk Drives .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
201  
201  
202  
Manual Rebuild – Rebuilding an Individual Drive  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Manual Rebuild – Batch Mode  
.
.
.
.
.
.
.
.
.
Exiting Dell Manager .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
202  
13 Troubleshooting  
Overview  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
204  
205  
BIOS Boot Error Messages .  
Other BIOS Error Messages  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
208  
209  
210  
Other Potential Problems .  
Audible Warnings . .  
.
.
.
.
.
Contents  
15  
14 Appendix A: SCSI Cables and Connectors  
15 Appendix B: Warranties and Return Policy  
Glossary  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
237  
247  
Index  
.
.
16  
Contents  
Figures  
Figure 2-1. Example of Disk Striping  
Figure 2-2. Example of Disk Mirroring  
Figure 2-3. Example of Disk Spanning .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
33  
34  
35  
Figure 2-4. Example of Parity .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
37  
Figure 3-1. RAID 0 Array  
Figure 3-2. RAID 1 Array  
Figure 3-3. RAID 5 Array  
Figure 3-4. RAID 10 Array  
Figure 3-5. RAID 50 Array  
.
.
.
.
.
.
.
.
.
.
.
.
.
44  
46  
48  
50  
52  
Figure 8-1. PERC 3/SC Card Layout  
103  
108  
110  
111  
112  
113  
119  
120  
123  
Figure 8-2. Example of SCSI Termination .  
Figure 8-3. Termination of Internal SCSI Disk Arrays  
Figure 8-4. PCI Slots on Motherboard .  
Figure 8-5. Installation of PERC 3/SC Card into Motherboard .  
.
.
.
.
.
.
.
Figure 8-6. Connecting SCSI Cables .  
Figure 9-1. PERC 3/DC Card Layout  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Figure 9-2. PERC 3/DCL Card Layout .  
Figure 9-3. J11 Serial Port Pinout  
.
.
Figure 9-4. Termination of Internal SCSI Disk Arrays for PERC 3/DC and  
3/DCL  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
126  
127  
129  
135  
139  
141  
142  
143  
144  
145  
Figure 9-5. Installation of the PERC 3/DC and 3/DCL  
.
.
.
.
.
Figure 9-6. Connecting SCSI Cables to PERC 3/DC and 3/DCL  
Figure 10-1. PERC 3/QC Card Layout .  
Figure 10-2. J14 Serial Port Diagram .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Figure 10-3. Example of Termination  
.
Figure 10-4. Termination of Internal SCSI Disk Arrays  
Figure 10-5. PCI Slots on Motherboard  
Figure 10-6. Installation of the PERC 3/QC Card  
Figure 10-7. Connecting SCSI Cables  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Figures  
17  
18  
Figures  
Tables  
Table 1-1. Maximum Cable Length for SCSI Standards  
Table 1-2. SCSI Bus Widths and Maximum Throughput  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
26  
27  
36  
36  
39  
40  
42  
43  
45  
47  
49  
51  
54  
55  
56  
56  
57  
57  
58  
61  
64  
65  
66  
67  
67  
68  
68  
71  
74  
75  
Table 2-1. Spanning for RAID 10 and RAID 50  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Table 2-2. Types of Parity  
Table 2-3. SCSI Drive States  
Table 2-4. Logical Drive States  
Table 3-1. RAID Levels  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Table 3-2. RAID 0 Overview .  
Table 3-3. RAID 1 Overview .  
Table 3-4. RAID 5 Overview .  
Table 3-5. RAID 10 Overview  
Table 3-6. RAID 50 Overview  
Table 4-1. Configuration Features  
Table 4-2. Configuration on Disk Features .  
Table 4-3. Hardware Architecture Features  
Table 4-4. Array Performance Features  
.
.
Table 4-5. PERC 3/SC Fault Tolerance Features .  
Table 4-6. Software Utilities Features  
Table 4-7. PERC 3/SC Specifications  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Table 4-8. SCSI Firmware Support  
Table 5-1. Configuration Features  
.
.
.
Table 5-2. Configuration on Disk Features .  
Table 5-3. Hardware Architecture Features  
Table 5-4. Array Performance Features  
.
.
.
.
.
.
Table 5-5. Fault Tolerance Features  
Table 5-6. Software Utilities .  
.
.
.
.
.
.
Table 5-7. PERC 3/DC and PERC 3/DCL Specifications  
Table 5-8. SCSI Firmware  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Table 6-1. Configuration Features  
Table 6-2. Configuration on Disk Features .  
Tables  
19  
Table 6-3. Hardware Architecture Features .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
76  
76  
Table 6-4. Array Performance Features  
.
.
.
.
.
.
.
.
.
.
Table 6-5. Fault Tolerance Features  
Table 6-6. Software Utilities  
.
.
.
.
.
.
77  
.
.
.
.
78  
Table 6-7. PERC 3/QC Specifications .  
Table 6-8. SCSI Firmware  
78  
.
.
.
.
.
.
81  
Table 7-1. Current Configuration for SCSI Channel 0 .  
Table 7-2. Current Configuration for SCSI Channel 1 .  
Table 7-3. Current Configuration for SCSI Channel 2 .  
Table 7-4. Current Configuration for SCSI Channel 3 .  
84  
85  
86  
87  
Table 7-5. Logical Drive Configuration  
Table 7-6. Physical Device Layout  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
87  
.
.
89  
Table 7-7. Drives and Capacities for Each RAID Level  
94  
Table 7-8. Fault Tolerance Features for RAID Levels 0, 1 and 5  
Table 7-9. Performance Characteristics for Each RAID Level .  
94  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
95  
Table 7-10. Physical Drives Required for Each RAID Level .  
Table 7-11. Factors to Consider for Array Configuration  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
95  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
97  
Table 7-12. Array Configuration Planner  
Table 8-1. PERC 3/SC Jumpers  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
98  
.
.
.
.
105  
105  
106  
106  
107  
107  
108  
115  
121  
122  
122  
123  
123  
Table 8-2. J1 Termination Enable Settings  
Table 8-3. J9 I2C Interface Connector Pinout  
Table 8-4. J5 Serial Port Pinout  
Table 8-5. J8 Hard Disk LED  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Table 8-6. J15 RUBI Slot Interrupt Steering .  
Table 8-7. J16, J17 RUBI Slot Interrupt Steering  
Table 8-8. Target IDs .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Table 9-1. Jumper Settings .  
Table 9-2. J2 and J3 Termination Enable Settings  
Table 9-3. J9 Onboard BIOS Enable Settings .  
.
.
.
.
.
.
Table 9-4. J10 NVRAM Clear Pinout  
Table 9-5. J11 Serial Port Pinout  
.
.
.
.
.
.
.
.
.
.
20  
Tables  
Table 9-6. J13 Dirty Cache LED Pinout  
Table 9-7. J14 SCSI Activity Pinout .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
124  
124  
124  
125  
128  
137  
138  
138  
139  
139  
140  
140  
146  
151  
152  
153  
155  
156  
157  
157  
161  
161  
182  
188  
189  
190  
191  
191  
192  
193  
193  
.
Table 9-8. J16 and J18 TERMPWR Enable Settings  
Table 9-9. J17 I2C Connector Pinout  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Table 9-10. Target IDs for SCSI Devices  
Table 10-1. PERC 3/QC Jumper Settings  
Table 10-2. J2, J3, J5 and J7 Termination Enable Pinout  
Table 10-3. J9, J10, J11 and J12 TERMPWR Enable Pinout  
Table 10-4. J14 Serial Port Pinout  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Table 10-5. J17 Dirty Cache LED Pinout  
Table 10-6. J19 Onboard BIOS Enable Settings  
Table 10-7. J23 External Battery Pinout  
Table 10-8. Target IDs  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Table 11-1. BIOS Configuration Utility Menu Options  
Table 11-2. Configuration Menu Options .  
Table 11-3. Adapter Menu Options .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Table 11-4. Logical Drive Menu Options .  
Table 11-5. Physical Drive Menu Options  
Table 11-6. SCSI Channel Menu Options  
Table 11-7. Battery Information Menu Options  
Table 11-8. PERC 3/QC and PERC 3/DC Default Settings  
Table 11-9. PERC 3/SC and PERC 3/DCL Default Settings  
Table 11-10. Rebuild Types  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Table 12-1. Command Used to Start Dell Manager  
Table 12-2. Management Menu Options  
Table 12-3. Adapter Menu Options  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Table 12-4. Logical Drive Menu Options .  
Table 12-5. Logical Drive Default Settings .  
Table 12-6. Physical Drive Menu Options  
Table 12-7. SCSI Channel Options .  
Table 12-8. Battery Backup Menu Items  
.
.
.
.
.
.
Tables  
21  
Table 12-9. Rebuild Types  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
201  
204  
205  
208  
209  
210  
Table 13-1. General Problems and Suggested Solutions .  
Table 13-2. BIOS Boot Error Messages  
Table 13-3. Other BIOS Error Messages  
Table 13-4. Other Potential Problems  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Table 13-5. Audible Warnings  
.
.
.
.
22  
Tables  
SE CT ION 1  
Overview  
PERC 3 Overview  
PERC 3 Features  
PERC 3 Overview  
The Dell™ PowerEdge™ Expandable RAID Controller 3 (PERC 3) PCI  
card is a high-performance, intelligent peripheral component interconnect  
(PCI)-to-SCSI host adapter with RAID control capabilities. PERC 3  
provides system availability, high performance, and fault-tolerant disk  
subsystem management. PERC 3 is an ideal RAID solution for the internal  
storage of Dell’s workgroup, departmental, and enterprise systems. PERC 3  
offers a cost-effective way to implement RAID in a server.  
The PERC 3/QC low voltage differential (LVD) PCI RAID card provides  
four SCSI channels. The PERC 3/DC and PERC 3/DCL LVD PCI RAID  
card provides two SCSI channels. The PERC 3/SC LVD PCI RAID card  
provides one SCSI channel.  
NOTE: The information  
in this document applies  
to PERC 3/QC, PERC  
3/DC, PERC 3/DCL and  
PERC 3/SC.  
With LVD, you can use cables up to 25 meters long. Throughput on each  
SCSI channel can be as high as 160 MB/s. PERC 3 supports both an LVD  
SCSI bus and a single-ended SCSI bus.  
PERC 3 64-bit LVD is a high-performance, intelligent PCI-to-SCSI host  
adapter with RAID control capabilities. PERC 3 64-bit LVD requires no  
special motherboard PCI expansion slot.  
PERC 3 Features  
The PERC 3 features include:  
•
•
Wide Ultra3 LVD SCSI performance of up to 160 MB/s  
PERC 3/QC and PERC 3/DC support 128 MB of synchronous  
dynamic random access memory (SDRAM.)  
•
•
•
•
The PERC 3/DCL supports 64 MB of SDRAM.  
The PERC 3/SC supports 32 MB of SDRAM.  
64-bit PCI host interface (only on PERC 3/QC and PERC 3/DC)  
The PERC 3/QC, PERC 3/DC and PERC 3/DCL have an on-board  
i960RN processor to improve controller performance and offload host  
Central Processing Unit (CPU.)  
•
The PERC 3/SC has a on-board i960RM processor to improve  
controller performance and offload host CPU.  
24  
Overview  
•
•
Two internal and four external connectors for the PERC 3/QC, two  
internal and two external connectors for the PERC 3/DC and PERC  
3/DCL, and one internal and one external connector for the PERC  
3/SC  
RAID levels 0 (striping), 1 (mirroring), 5 (distributed parity), 10  
(combination of striping and mirroring), and 50 (combination of  
striping and distributed parity)  
•
•
Advanced array configuration and management utilities  
Battery backup for up to 72 hours for the PERC 3/QC and PERC  
3/DC  
•
Up to 12 SCSI drives per channel using the Dell PowerVault 21xS  
storage system or 14 SCSI drives using the Dell PowerVault 22xS  
NOTE: Clustering is not  
supported by PERC 3/QC  
or PERC 3/DCL.  
SCSI Channels  
The PERC 3/QC controller card includes four SCSI channels. There are two  
QLogic®12160 chips; each chip controls two SCSI channels.  
The PERC 3/DC and PERC 3/DCL controller cards include two SCSI  
channels. There is one QLogic 12160 chip, which controls two SCSI  
channels.  
The PERC 3/SC controller card has one SCSI channel. There is one QLogic  
10160 chip.  
Non-volatile Random Access Memory (NVRAM) and Flash Read-  
only Memory (ROM)  
A 32 KB x 8 NVRAM stores RAID system configuration information. The  
PERC 3 firmware is stored in flash ROM for easy upgrade.  
SCSI Connectors  
PERC 3/QC has four ultra-high-density 68-pin external connectors for the  
external storage subsystem and two high-density 68-pin internal connectors.  
PERC 3/DC and PERC 3/DCL have two ultra-high-density 68-pin external  
connectors for the external storage subsystem and two high-density 68-pin  
internal connectors.  
Overview  
25  
PERC 3/SC has one ultra-high-density 68-pin external connector for the  
external storage subsystem and one high-density 68-pin internal connector.  
Single-Ended and LVD SCSI Buses  
The SCSI standard defines two electrical buses: a single-ended bus and an  
LVD bus. PERC 3 supports both standards.  
Maximum Cable Length for SCSI Standards  
Table 1-1 displays the maximum length of cable that you can use for various  
SCSI standards.  
Table 1-1. Maximum Cable Length for SCSI Standards  
SCSI Standard  
Single-ended  
LVD  
Maximum # of  
Drives  
SCSI I  
6 m  
6 m  
6 m  
1.5 m  
3 m  
N/A  
1.5 m  
3 m  
N/A  
N/A  
N/A  
N/A  
12 m  
12 m  
12 m  
12 m  
12 m  
12 m  
12 m  
12 m  
25 m  
12 m  
25 m  
12 m  
7
Fast SCSI  
7
Fast Wide SCSI  
Ultra SCSI  
15  
7
Ultra SCSI  
3
Wide Ultra SCSI  
Wide Ultra SCSI  
Wide Ultra SCSI  
Ultra 2 SCSI  
15  
7
3
1
Ultra 2 SCSI  
7
Wide Ultra 2 SCSI  
Wide Ultra 2 SCSI  
1
15  
SCSI Bus Widths and Maximum Throughput  
Table 1-2 displays the SCSI bus widths and maximum throughput for each  
of the SCSI standards.  
26  
Overview  
Table 1-2. SCSI Bus Widths and Maximum Throughput  
SCSI Standard  
SCSI I  
SCSI Bus Width  
8 bits  
SCSI Throughput  
5 MB/s  
Fast SCSI  
8 bits  
10 MB/s  
Fast Wide SCSI  
Ultra SCSI  
16 bits  
8 bits  
20 MB/s  
20 MB/s  
Wide Ultra SCSI  
Ultra 2 SCSI  
Wide Ultra 2 SCSI  
160M SCSI  
16 bits  
8 bits  
40 MB/s  
40 MB/s  
16 bits  
8 bits  
80 MB/s  
80 MB/s  
Wide 160M SCSI  
16 bits  
160 MB/s  
Operating System Support  
PERC 3 supports the following operating systems:  
•
Microsoft® Windows® 2000: Server, Advanced Server, Small Business  
Server  
•
Microsoft Windows NT® 4: Server, Terminal Server, Enterprise Server,  
Small Business Server  
•
•
•
•
•
Microsoft NET: Server, Advanced Server  
Microsoft SBS 2000  
Novell® NetWare® 5.x, 6.x  
Red Hat Linux 7.x  
DOS, including ASPI® support  
Overview  
27  
28  
Overview  
SE CT ION 2  
Introduction to RAID  
RAID Definition  
PERC 3 Host-Based RAID Solution  
Components and Features  
RAID Definition  
RAID is an array of multiple independent hard disk drives that provides  
high performance and fault tolerance. The RAID array appears to the host  
computer as a single storage unit or as multiple logical units. Input/output  
(I/O) improves because several disks can be accessed simultaneously.  
Although hard drive capabilities have improved drastically, actual  
performance has improved only three to four times in the last decade. In  
comparison, computing performance has improved over 50 times during the  
same time period. Implementing RAID improves the performance of the  
disk subsystem.  
RAID systems also improve data storage availability and fault tolerance.  
Data loss caused by a hard drive failure can be recovered by reconstructing  
missing data from the remaining data and parity drives.  
PERC 3 Host-Based RAID Solution  
PERC 3 is a peripheral component interconnect (PCI) adapter card that is  
installed in any available PCI expansion slot in a host system. It is a host-  
based RAID solution, meaning that the RAID controller puts all of the  
RAID intelligence on an adapter card that is installed in a network server.  
The array controller resides on the bus (for example, a PCI or EISA bus) in  
the host computer and has its own central processing unit (CPU) to  
generate the parity and handle other RAID functions. A bus-based  
controller can transfer data at the speed of the host bus (PCI, ISA, EISA,  
VL-Bus), but is limited to the bus it is designed for. PERC 3 resides on a  
PCI bus, which can handle data transfer at up to 528 MB/s. With PERC 3,  
each channel can handle data transfer rates up to 160 MB/s per SCSI  
channel. The available sequential data transfer rate is determined by the  
following factors:  
•
•
The sustained data transfer rate on the motherboard PCI bus  
The sustained data transfer rate on the i960RN (i960RM for the PERC  
3/SC) PCI to PCI bridge  
•
•
The sustained data transfer rate of the SCSI controller  
The sustained data transfer rate of the SCSI devices  
30  
Introduction to RAID  
•
The number of SCSI channels and SCSI hard drives  
Components and Features  
RAID versions, or levels, are specifications that describe a system for  
ensuring the availability and stability of data stored on large disk  
subsystems. A RAID system can be implemented in a number of different  
levels). PERC 3 supports RAID levels 0, 1, 5, 10 (1+0), and 50 (5+0).  
Physical Array  
A RAID array is a collection of physical disk drives governed by the RAID  
management software. A RAID array appears to the host computer as one or  
more logical drives.  
Logical Drive  
A logical drive is a partition in a physical array of disks that is made up of  
contiguous data segments on the physical disks. A logical drive can consist  
of an entire physical array, more than one entire physical array, a part of an  
array, parts of more than one array, or a combination of any two of these  
conditions.  
Fault Tolerance  
Fault tolerance is the capability of the subsystem to undergo a single failure  
without compromising data integrity, and processing capability. The RAID  
controller provides this support through redundant arrays in RAID levels 1,  
5, 10 and 50. The system can still work properly even with a single disk  
failure in an array, through performance can be degraded to some extent.  
Fault tolerance is often associated with system availability (high mean time  
between failure, MTFB) because it allows the system to be available during  
the failures. However, this means it is also important for the system to be  
available during the repair of the problem. To make this possible, PERC 3  
supports hot spare disks, and the auto-rebuild feature.  
Introduction to RAID  
31  
A hot spare is an unused online disk that, in case of a disk failure in a  
redundant RAID array, can be used to rebuild the data and re-establish  
redundancy. After the hot spare is automatically moved into the RAID  
subsystem, the failed drive is automatically rebuilt on the spare drive. The  
RAID disk array continues to handle requests while the rebuild occurs.  
Auto-rebuild allows a failed drive to be replaced and automatically rebuilt  
by “hot-swapping” the drive in the same drive bay. The RAID disk array  
continues to handle requests while the rebuild occurs.  
Consistency Check  
In RAID, check consistency verifies the correctness of redundant data in an  
array. For example, in a system with parity, checking consistency means  
computing the data on one drive and comparing the results to the contents  
of the parity drive.  
Disk Striping  
Disk striping allows you to write data across multiple disk drives instead of  
just one disk drive. Disk striping involves partitioning each drive storage  
space into stripes that can vary in size from 2 KB to 128 KB. These stripes  
are interleaved in a repeated sequential manner. The combined storage  
space is composed of stripes from each drive. PERC 3 supports stripe sizes  
of 2 KB, 4 KB, 8 KB, 16 KB, 32 KB, 64 KB, and 128 KB.  
For example, in a four-disk system using only disk striping (as in RAID level  
0), segment 1 is written to disk 1, segment 2 is written to disk 2, and so on.  
Disk striping enhances performance because multiple drives are accessed  
simultaneously, but disk striping does not provide data redundancy.  
Figure 2-1 shows an example of disk striping.  
32  
Introduction to RAID  
Figure 2-1. Example of Disk Striping  
Segment 1  
Segment 5  
Segment 9  
Segment 2  
Segment 6  
Segment 10  
Segment 3  
Segment 7  
Segment 11  
Segment 4  
Segment 8  
Segment 12  
Stripe Width  
Stripe width is the number of disks involved in an array where striping is  
implemented. For example, a four-disk array with disk striping has a stripe  
width of four.  
Stripe Size  
The stripe size is the length of the interleaved data segments that PERC 3  
writes across multiple drives. PERC 3 supports stripe sizes of 2 KB, 4 KB,  
8 KB, 16 KB, 32 KB, 64 KB, or 128 KB.  
Introduction to RAID  
33  
Disk Mirroring  
With mirroring (used in RAID 1), data written to one disk is simultaneously  
written to another disk. If one disk fails, the contents of the other disk can  
be used to run the system and reconstruct the failed disk. The primary  
advantage of disk mirroring is that it provides 100% data redundancy.  
Because the contents of the disk are completely written to a second disk, it  
does not matter if one of the disks fails. Both disks contain the same data at  
all times. Either drive can act as the operational drive.  
Disk mirroring provides 100% redundancy, but is expensive because each  
drive in the system must be duplicated.  
Figure 2-2. Example of Disk Mirroring  
Segment 1  
Segment 2  
Segment 3  
Segment 4  
Segment 1 Duplicated  
Segment 2 Duplicated  
Segment 3 Duplicated  
Segment 4 Duplicated  
34  
Introduction to RAID  
Disk Spanning  
Disk spanning allows multiple physical drives to function like one big drive.  
Spanning overcomes lack of disk space and simplifies storage management  
by combining existing resources or adding relatively inexpensive resources.  
For example, four 20 GB drives can be combined to appear to the operating  
system as a single 80 GB drive.  
Spanning alone does not provide reliability or performance enhancements.  
Spanned logical drives must have the same stripe size and must be  
contiguous. In the following graphic, RAID 1 array is turned into a RAID 10  
array.  
Figure 2-3. Example of Disk Spanning  
60 GByte  
60 GByte  
60 GByte  
60 GByte  
Can Be Accessed as  
One 120 GByte Drive  
Can Be Accessed as  
One 120 GByte Drive  
Introduction to RAID  
35  
Spanning for RAID 10 or RAID 50  
Table 2-1 describes how to configure RAID 10 and RAID 50 by spanning.  
NOTE: Spanning two  
contiguous RAID 0  
Table 2-1. Spanning for RAID 10 and RAID 50  
logical drives does not  
produce a new RAID level  
or add fault tolerance. It  
does increase the size of  
the logical volume and  
improves performance by  
doubling the number of  
spindles.  
Level  
Description  
10  
Configure RAID 10 by spanning two contiguous RAID 1 logical drives.  
The RAID 1 logical drives must have the same stripe size.  
NOTE: Refer to Chapter 11 "PERC 3 BIOS Configuration  
Utility" for the configuration procedure for spanning RAID 1  
logical drives.  
50  
Configure RAID 50 by spanning two contiguous RAID 5 logical drives.  
The RAID 5 logical drives must have the same stripe size.  
NOTE: Refer to Chapter 11 "PERC 3 BIOS Configuration  
Utility" for the configuration procedure for spanning RAID 5  
logical drives.  
Parity  
Parity generates a set of redundancy data from two or more parent data sets.  
The redundancy data can be used to reconstruct one of the parent data sets.  
Parity data does not fully duplicate the parent data sets. In RAID, this  
method is applied to entire drives or stripes across all disk drives in an array.  
The types of parity are shown in Table 2-2.  
Table 2-2. Types of Parity  
Parity Type  
Description  
Dedicated  
The parity of the data on two or more disk drives is stored on an  
additional disk.  
Distributed  
The parity data is distributed across all drives in the system.  
If a single disk drive fails, it can be rebuilt from the parity and the data on  
the remaining drives. RAID level 5 combines distributed parity with disk  
striping. Parity provides redundancy for one drive failure without  
duplicating the contents of entire disk drives, but parity generation can slow  
the write process. A dedicated parity scheme during normal read/write  
operations is shown in Figure 2-4.  
36  
Introduction to RAID  
Figure 2-4. Example of Parity  
Parity  
Generator  
Hot Spares  
A hot spare is an extra, unused disk drive that is part of the disk subsystem.  
It is usually in standby mode, ready for service if a drive fails. Hot spares  
permit you to replace failed drives without system shutdown or user  
intervention.  
PERC 3 implements automatic and transparent rebuilds using hot spare  
drives, providing a high degree of fault tolerance and zero downtime. The  
PERC 3 RAID management software allows you to specify physical drives as  
hot spares. When a hot spare is needed, PERC 3 assigns the hot spare that  
has a capacity closest to and at least as great as that of the failed drive to  
take the place of the failed drive.  
Introduction to RAID  
37  
NOTE: Refer to Chapter  
11 "PERC 3 BIOS  
Configuration Utility" for  
the procedures used to  
designate a drive as a hot  
spare.  
There are two types of hot spares:  
•
•
Global Hot Spare  
Dedicated Hot Spare  
Global Hot Spare  
A global hot spare drive can be used to replace any failed drive in a  
redundant array as long as its capacity is equal to or larger than the coerced  
capacity of the failed drive. A global hot spare defined on any channel  
should be available to replace a failed drive on both channels.  
Dedicated Hot Spare  
A dedicated hot spare can be used to replace a failed drive only in a selected  
array. One or more drives can be designated as member of a spare drive pool;  
the most suitable drive from the pool is selected for fail over. A dedicated  
hot spare is used before one from the global hot spare pool.  
NOTE: Refer to Chapter  
11"PERC 3 BIOS  
Configuration Utility" for  
the procedures used to  
designate a drive as a hot  
spare.  
Hot spare drives can be located on any RAID channel. Standby hot spares  
(not being used in RAID array) are polled every 60 seconds at a minimum,  
and their status made available in the array management software. PERC 3  
offers the ability to rebuild with a disk that is in a system, but not initially  
set to be a hot spare.  
Observe the following parameters when using hot spares:  
•
•
•
Hot spares are used only in arrays with redundancy, for example, RAID  
levels 1, 5, 10, and 50.  
A hot spare connected to a specific PERC 3 controller can be used to  
rebuild a drive that is connected to the same controller only.  
You must assign the hot spare to one or more drives through the  
controller’s BIOS or use array management software to place it in the  
hot spare pool.  
•
A hot spare must have free space equal to or greater than the drive it  
would replace. For example, to replace an 18 GB drive, the hot spare  
must be 18 GB or larger.  
Disk Rebuilds  
You can rebuild a disk drive by recreating the data that had been stored on  
the drive before the drive failed.  
38  
Introduction to RAID  
Rebuilding can be done only in arrays with data redundancy, which includes  
as RAID 1, 5, 10 and 50.  
A hot spare can be used to rebuild disk drives in RAID 1, 5, 10, or 50  
systems. If a hot spare is not available, the failed disk drive must be replaced  
with a new disk drive so that the data on the failed drive can be rebuilt.  
Using hot spares, PERC 3 can automatically and transparently rebuild failed  
drives with user-defined rebuild rates. If a hot spare is available, the rebuild  
can start automatically when a drive fails. PERC 3 automatically restarts the  
system and the rebuild if the system goes down during a rebuild.  
Rebuild Rate  
The rebuild rate is the percentage of the compute cycles dedicated to  
rebuilding failed drives. A rebuild rate of 100 percent means the system is  
totally dedicated to rebuilding the failed drive.  
The PERC 3 rebuild rate can be configured between 0% and 100%. At 0%,  
the rebuild is done only if the system is not doing anything else. At 100%,  
the rebuild has a higher priority than any other system activity. The default  
rebuild rate is 30%.  
Hot Swap  
A hot swap is the manual replacement of a defective physical disk unit while  
the computer is still running. When a new drive has been installed, a  
rebuild will occur automatically if it is placed in the same drive bay as the  
failed drive it is replacing. PERC 3 can be configured to detect the new disks  
and to rebuild the contents of the disk drive automatically.  
SCSI Drive States  
The SCSI drive states are listed in Table 2-3.  
Table 2-3. SCSI Drive States  
State  
Description  
Online  
(ONLIN)  
The drive is working normally and is a part of a configured  
logical drive.  
Introduction to RAID  
39  
Table 2-3. SCSI Drive States (continued)  
State  
Description  
Ready  
The drive is functioning normally but is not part of a  
configured logical drive and is not designated as a hot spare.  
(READY)  
Hot Spare  
(HOTSP)  
Fail (FAIL)  
Rebuild (REB)  
The drive is powered up and ready for use as a spare in case an  
online drive fails.  
A fault has occurred in the drive, placing it out of service.  
The drive is being rebuilt with data from a failed drive.  
Logical Drive States  
The logical drive states are listed in Table 2-4.  
Table 2-4. Logical Drive States  
State  
Description  
Optimal  
The drive operating condition is good. All configured drives are  
online.  
Degraded  
The drive operating condition is not optimal. One of the  
configured drives has failed or is offline.  
Failed  
The drive has failed.  
Offline  
The drive is not available to the RAID controller.  
Enclosure Management  
Enclosure management is the intelligent monitoring of the disk subsystem  
by software and/or hardware. The disk subsystem can be part of the host  
computer or separate from it. Enclosure management helps you stay  
informed of events in the disk subsystem, such as a drive or power supply  
failure. Enclosure management increases the fault tolerance of the disk  
subsystem.  
40  
Introduction to RAID  
SE CT ION 3  
RAID Levels  
Overview  
Selecting a RAID Level  
RAID 0  
RAID 1  
RAID 5  
RAID 10  
RAID 50  
Overview  
There are six official RAID levels (RAID 0 through RAID 5.) PERC 3  
supports RAID levels 0, 1, 5, 10, and 50. The RAID levels that PERC 3  
supports are shown in Table 3-1.  
Table 3-1. RAID Levels  
RAID  
Level  
Type  
0
Standard  
1
Standard  
5
Standard  
1+0  
5+0  
Combination of RAID 0 and RAID 1  
Combination of RAID 0 and RAID 5  
PERC 3 also supports independent drives (configured as RAID 0.)  
Selecting a RAID Level  
To ensure the best performance, you should select the optimal RAID level  
when you create a system drive. The optimal RAID level for your disk array  
depends on a number of factors:  
•
•
•
•
The number of physical drives in the disk array  
The capacity of the physical drives in the array  
The need for data redundancy  
The disk performance requirements  
42  
RAID Levels  
RAID 0  
RAID 0 provides disk striping across all drives in the RAID subsystem. RAID  
0 does not provide any data redundancy, but does offer the best  
performance of any RAID level. RAID 0 breaks up data into smaller blocks  
and then writes a block to each drive in the array. The size of each block is  
determined by the stripe size parameter, set during the creation of the RAID  
set. RAID 0 offers high bandwidth.  
By breaking up a large file into smaller blocks, PERC 3 can use several drives  
to read or write the file faster. RAID 0 involves no parity calculations to  
complicate the write operation. This makes RAID 0 ideal for applications  
that require high bandwidth but do not require fault tolerance. RAID 0 is  
also used to denote an “independent” or single drive.  
Table 3-2 provides an overview of RAID 0.  
Table 3-2. RAID 0 Overview  
Uses  
RAID 0 provides high data throughput, especially for  
large files. Any environment that does not require  
fault tolerance.  
Strong Points  
Weak Points  
Drives  
Provides increased data throughput for large files. No  
capacity loss penalty for parity.  
Does not provide fault tolerance. All data lost if any  
drive fails.  
1 to 32  
RAID Levels  
43  
Figure 3-1 displays an example of a RAID 0 array.  
Figure 3-1. RAID 0 Array  
Segment 1  
Segment 5  
Segment 7  
Segment 2  
Segment 6  
Segment 10  
Segment 3  
Segment 7  
Segment 11  
Segment 4  
Segment 8  
Segment 12  
44  
RAID Levels  
RAID 1  
In RAID 1, the RAID controller duplicates all data from one drive to a  
second drive. RAID 1 provides complete data redundancy, but at the cost of  
doubling the required data storage capacity, as shown in Figure 3-2. Table 3-  
3 provides an overview of RAID 1.  
Table 3-3. RAID 1 Overview  
Uses  
Use RAID 1 for small databases or any other  
environment that requires fault tolerance but small  
capacity.  
Strong Points  
RAID 1 provides complete data redundancy. RAID 1 is  
ideal for any application that requires fault tolerance  
and minimal capacity.  
Weak Points  
Drives  
RAID 1 requires twice as many disk drives.  
Performance is impaired during drive rebuilds.  
2
RAID Levels  
45  
Figure 3-2. RAID 1 Array  
Segment 1  
Segment 2  
Segment 3  
Segment 4  
Segment 1 Duplicated  
Segment 2 Duplicated  
Segment 3 Duplicated  
Segment 4 Duplicated  
RAID 5  
RAID 5 includes disk striping at the block level and parity. In RAID 5, the  
parity information is written to several drives. RAID 5 is best suited for  
networks that perform a lot of small input/output (I/O) transactions  
simultaneously.  
RAID 5 addresses the bottleneck issue for random I/O operations. Because  
each drive contains both data and parity numerous writes can take place  
concurrently. In addition, robust caching algorithms and hardware based  
exclusive-or assist make RAID 5 performance exceptional in many different  
environments.  
Table 3-4 provides an overview of RAID 5.  
46  
RAID Levels  
Table 3-4. RAID 5 Overview  
Uses  
RAID 5 provides high data throughput, especially for  
large files. Use RAID 5 for transaction processing  
applications because each drive can read and write  
independently. If a drive fails, PERC 3 uses the parity  
drive to recreate all missing information. Use also for  
office automation and online customer service that  
requires fault tolerance. Use for any application that  
has high read request rates but low write request rates.  
Strong Points  
Weak Points  
Provides data redundancy and good performance in  
most environments  
Disk drive performance will be reduced if a drive is  
being rebuilt. Environments with few processes do not  
perform as well because the RAID overhead is not  
offset by the performance gains in handling  
simultaneous processes.  
Drives  
3 to 32  
RAID Levels  
47  
Figure 3-3 displays an example of a RAID 5 array.  
Figure 3-3. RAID 5 Array  
Data Flow  
Disk 1  
Disk 2  
Disk 3  
Disk 4  
Disk 5  
Disk 6  
Segment 1  
Segment 7  
Parity (9–12)  
Segment 2  
Segment 8  
Segment 3  
Segment 9  
Parity (5–8)  
Segment 4  
Segment 10  
Segment 5  
Segment 11  
Parity (1–4)  
Segment 6  
Segment 12  
Note: Parity is distributed across drives in the array.  
RAID 10  
RAID 10 is a combination of RAID 0 and RAID 1. RAID 10 has mirrored  
drives. RAID 10 breaks up data into smaller blocks, and then mirrors the  
blocks of data to each RAID 1 RAID set. Each RAID 1 RAID set then  
duplicates its data to its other drive. The size of each block is determined by  
the stripe size parameter, which is set during the creation of the RAID set.  
RAID 10 can sustain one to four drive failures while maintaining data  
integrity if each failed disk is in a different RAID 1 array.  
Table 3-5 provides an overview of RAID 10.  
48  
RAID Levels  
Table 3-5. RAID 10 Overview  
Uses  
RAID 10 works best for data storage that needs 100%  
redundancy of mirrored arrays and that also needs the  
enhanced I/O performance of RAID 0 (striped arrays.)  
RAID 10 works well for medium-sized databases or  
any environment that requires a higher degree of fault  
tolerance and moderate to medium capacity.  
Strong Points  
Weak Points  
Drives  
RAID 10 provides both high data transfer rates and  
complete data redundancy.  
RAID 10 requires twice as many drives as all other  
RAID levels except RAID 1.  
2n, where n is greater than 1. The maximum number  
of drives is 16.  
RAID Levels  
49  
Figure 3-4. RAID 10 Array  
Data Flow  
RAID 1  
RAID 1  
D
Disk 1  
Disk 2  
Disk 3  
Disk 4  
Segment 1  
Segment 1  
Segment 2  
Segment 2  
Segment 3  
Segment 5  
Segment 3  
Segment 5  
Segment 4  
Segment 6  
Segment 4  
Segment 6  
RAID 0  
RAID 50  
RAID 50 provides the features of both RAID 0 and RAID 5. RAID 50  
includes both parity and disk striping across multiple drives. RAID 50 is best  
implemented on two RAID 5 disk arrays with data striped across both disk  
arrays.  
RAID 50 breaks up data into smaller blocks, and then stripes the blocks of  
data to each RAID 5 array. RAID 5 breaks up data into smaller blocks,  
calculates parity by performing an exclusive-or on the blocks, and then  
writes the blocks of data and parity to each drive in the array. The size of  
each block is determined by the stripe size parameter, which is set during  
the creation of the RAID set.  
50  
RAID Levels  
RAID 50 can sustain one to four drive failures while maintaining data  
integrity if each failed disk is in a different RAID 5 array.  
Table 3-6 provides an overview of RAID 50.  
Table 3-6. RAID 50 Overview  
Uses  
RAID 50 works best when used with data that requires  
high reliability, high request rates, and high data  
transfer and medium to large capacity.  
Strong Points  
RAID 50 provides high data throughput, data  
redundancy, and very good performance.  
Weak Points  
Drives  
Requires 2 to 4 times as many parity drives as RAID 5.  
6 to 256 (Eight arrays x 32 stripes = 256 drives)  
RAID Levels  
51  
Figure 3-5 displays an example of a RAID 50 array.  
Figure 3-5. RAID 50 Array  
Data Flow  
RAID 5  
RAID 5  
Disk 4  
Disk 5  
Disk 6  
Disk 1  
Disk 2  
Disk 3  
Segment 1  
Segment 6  
Parity (9-10)  
Segment 2  
Parity (5-6)  
Segment 9  
Parity (1-2)  
Segment 5  
Segment 10  
Segment 3  
Segment 8  
Parity (11-12)  
Segment 4  
Parity (7-8)  
Segment 11  
Parity (3-4)  
Segment 7  
Segment 12  
RAID 0  
52  
RAID Levels  
SE CT ION 4  
PERC 3/SC Features  
Hardware Requirements  
Configuration Features  
Hardware Architecture Features  
Array Performance Features  
PERC 3/SC Fault Tolerance Features  
Software Utilities  
Operating System Software Drivers  
PERC 3/SC Specifications  
Hardware Requirements  
PERC 3/SC has one SCSI) channel that supports 160M and Wide SCSI at  
data transfer rates up to 160 MB/s. The SCSI channel supports up to 15  
Wide devices or up to seven narrow devices.  
PERC 3/SC can be installed in a Dell™ PowerEdge™ computer with a  
motherboard that has 5 V, 32- or 64-bit PCI expansion slots. The computer  
should have an Intel® Pentium® II or higher central processing unit (CPU),  
a diskette drive, a color monitor and video graphics adapter (VGA) card, and  
a keyboard. A mouse is recommended.  
Configuration Features  
Table 4-1 lists the configuration features for the PERC 3/SC controller.  
Table 4-1. Configuration Features  
Specification  
RAID levels  
Feature  
0, 1, 5, 10, and 50.  
SCSI channels  
1
Maximum number of drives per  
channel  
15  
Array interface to host  
Drive interface  
PCI 2.2  
Wide 160M  
32 MB  
Cache memory size  
Cache function  
Write-back, write-through, adaptive  
read-ahead, non-read-ahead, read-  
ahead  
Multiple logical drives/arrays per  
controller  
Up to 40 logical drives per controller  
Maximum number of PERC 3/SC  
controller per system  
Contact your Dell representative.  
Online capacity expansion  
Dedicated and pool hot spare  
Flashable firmware  
Yes  
Yes  
Yes  
Yes  
Hot swap devices supported  
54  
PERC 3/SC Features  
Table 4-1. Configuration Features (continued)  
Specification  
Feature  
Non-disk devices supported  
Mixed capacity hard disk drives  
Number of 16-bit internal connectors  
No  
Yes  
1
SMART Technology  
The PERC 3/SC self-monitoring analysis and reporting technology  
(SMART) detects predictable drive failures. SMART monitors the internal  
performance of all motors, heads, and drive electronics.  
Configuration on Disk Configuration  
Configuration on Disk (drive roaming) saves configuration information  
both in non-volatile random access memory (NVRAM) on PERC 3/SC and  
on the disk drives attached to PERC 3/SC. If the controller is replaced, the  
new controller detects the actual RAID configuration from the  
configuration information on the drives. This maintains the integrity of the  
data on each drive, even if the drives have changed their target ID.  
NOTE: Configuration on  
Disk does not work if you  
change both the adapter,  
and the SCSI connectors  
to different connectors on  
the new adapter. It will  
work only if you make one  
change at a time.  
Drive roaming is supported across channels on the same controller. Drive  
roaming across channels is not supported when cluster mode is enabled.  
Table 4-2 lists the Configuration on Disk features for the PERC 3/SC  
controller.  
Table 4-2. Configuration on Disk Features  
Specification  
Feature  
Support for hard disk drives with  
Yes  
capacities of more than 8 gigabytes (GB)  
Online RAID level migration  
RAID remapping  
Yes  
Yes  
No reboot necessary after expansion  
Yes (only if virtual sizing is  
enabled)  
More than 200 qtags per array  
Yes  
Yes  
Hardware clustering support on the board  
PERC 3/SC Features  
55  
Table 4-2. Configuration on Disk Features (continued)  
Specification  
Feature  
User-specified rebuild rate  
Yes  
Hardware Architecture Features  
The PERC 3/SC hardware architecture features are shown in Table 4-3.  
Table 4-3. Hardware Architecture Features  
Specification  
Processor  
Feature  
Intel i960RM 100MHz  
®
SCSI controller(s)  
One QLogic 10160 Single SCSI  
controller  
Size of flash ROM  
Amount of NVRAM  
1 MB  
32 KB  
Yes  
Hardware-exclusive OR (XOR)  
assistance  
Direct I/O  
Yes  
SCSI bus termination  
Active, single-ended or low voltage  
differential (LVD)  
Double-sided dual in-line memory  
modules (DIMMs)  
Yes  
Auxiliary TermPWR source  
Direct I/O bandwidth  
No  
132 MB/s  
Array Performance Features  
The PERC 3/SC array performance features are shown in Table 4-4.  
Table 4-4. Array Per formance Features  
Specification  
Feature  
132 MB/s  
160 MB/s  
Host data transfer rate  
Drive data transfer rate  
56  
PERC 3/SC Features  
Table 4-4. Array Per formance Features (continued)  
Specification  
Feature  
Maximum scatter/gathers  
Maximum size of I/O requests  
Maximum queue tags per drive  
Stripe sizes  
26 elements  
6.4 MB in 64 KB stripes  
As many as the drive can accept.  
2 KB, 4 KB, 8 KB, 16 KB, 32 KB,  
64 KB, or 128 KB  
Maximum number of concurrent  
commands  
255  
Support for multiple initiators  
Yes  
PERC 3/SC Fault Tolerance Features  
The PERC 3/SC fault tolerance features are shown in Table 4-5.  
Table 4-5. PERC 3/SC Fault Tolerance Features  
Specification  
Feature  
Yes  
Support for SMART  
Drive failure detection  
Drive rebuild using hot spares  
Automatic  
Automatic  
Yes  
Manual replacement of drives without system  
shutdown  
Parity generation and checking  
Yes  
Software Utilities  
The PERC 3/SC software utilities are shown in Table 4-6.  
Table 4-6. Software Utilities Features  
Specification  
Feature  
Yes  
Graphical user interface  
Management utility  
Yes  
PERC 3/SC Features  
57  
Table 4-6. Software Utilities Features (continued)  
Specification  
Feature  
Bootup configuration using the PERC BIOS  
configuration utility (Ctrl–M)  
Yes  
Online read, write, and cache policy switching  
Intranet support  
Yes  
Yes  
Operating System Software Drivers  
Operating System Drivers  
PERC 3/SC includes a disk operating system (DOS) software configuration  
utility and drivers for Microsoft® Windows NT® V4.0, Windows® 2000,  
Novell® NetWare® 5.x, 6.x, and Red Hat Linux 7.x.  
PERC 3/SC Specifications  
The PERC 3/SC specifications are shown in Table 4-7.  
Table 4-7. PERC 3/SC Specifications  
Parameter  
Card size  
Processor  
Specification  
Half-length PCI adapter card size (6.875" X 4.2")  
Intel i960RM™ 64-bit RISC processor @  
100 MHz  
Bus type  
PCI 2.2  
PCI controller  
Bus data transfer rate  
Intel i960RM  
Up to 132 MB/s at 32/33 MHz  
PERC 3/SC BIOS  
Basic input/output system  
(BIOS)  
Cache configuration  
Firmware  
32 MB  
1 MB × 8 flash ROM  
32 KB × 8 for storing RAID configuration  
Non-volatile random access  
memory (RAM)  
Operating voltage  
5 V, 3.3 V  
58  
PERC 3/SC Features  
Table 4-7. PERC 3/SC Specifications (continued)  
Parameter  
Specification  
SCSI controller  
One SCSI controller for 160M and Wide support  
Up to 160 MB/s per channel  
SCSI data transfer rate  
SCSI bus  
Low-voltage differential (LVD) or single-ended  
Active  
SCSI termination  
Termination disable  
Devices per SCSI channel  
SCSI device types  
RAID levels supported  
SCSI connectors  
Automatic through cable and device detection  
Up to 15 wide or seven narrow SCSI devices  
Synchronous or asynchronous  
0, 1, 5, 10, and 50  
One 68-pin internal high-density connectors for  
16-bit SCSI devices. One ultra-high density 68-  
pin external connectors for 160M and Wide  
SCSI.  
Serial port  
3-pin RS232C-compatible connector (for  
manufacturing use only)  
PCI Bridge/CPU  
PERC 3/SC uses the Intel i960RM PCI bridge with an embedded i960RM  
RISC processor running at 100 MHz. The RM bridge handles data transfers  
between the primary (host) PCI bus, the secondary PCI bus, cache memory,  
and the SCSI bus. The DMA controller supports chaining and unaligned  
data transfers. The embedded i960JT CPU directs all controller functions,  
including command processing, SCSI bus transfers, RAID processing, drive  
rebuilding, cache management, and error recovery.  
Cache Memory  
32 MB of PERC 3/SC cache memory resides in a memory bank. PERC 3/SC  
supports write-through or write-back caching, selectable for each logical  
drive. To improve performance in sequential disk accesses, the PERC 3/SC  
controller uses read-ahead caching by default. You can disable read-ahead  
caching.  
PERC 3/SC Features  
59  
PERC 3/SC BIOS  
The BIOS resides on a 1 MB × 8 flash ROM for easy upgrade. The PERC  
3/SC BIOS supports INT 13h calls to boot DOS without special software or  
device drivers. The PERC 3/SC BIOS provides an extensive setup utility  
that can be accessed by pressing <Ctrl> <M> at BIOS initialization. The  
PERC 3/SC BIOS Configuration Utility is described in Chapter 11.  
Onboard Speaker  
The PERC 3/SC controller has an onboard tone generator for audible  
warnings when system errors occur. Audible warnings can be generated  
through this speaker. The audible warnings are listed in Chapter 13.  
Serial Port  
PERC 3/SC includes a 3-pin RS232C-compatible serial port connector. This  
is for manufacturing use only.  
SCSI Bus  
PERC 3/SC has one Wide 160M SCSI channels that support both LVD and  
single-ended devices with active termination. Synchronous and  
asynchronous devices are supported. PERC 3/SC provides automatic  
termination disable via cable detection. The SCSI channel supports up to  
15 wide or seven non-wide SCSI devices at speeds up to 160 MB/s.  
SCSI Connectors  
PERC 3/SC has two types of SCSI connectors:  
•
•
One 68-pin high density internal connector  
One 68-pin external ultra-high-density connector  
Only one connector type can be used for the SCSI channel at any one time.  
SCSI Termination  
PERC 3/SC uses active termination on the SCSI bus conforming to SCSI-2  
and SCSI-3 specifications. Termination enable/disable is automatic through  
cable detection.  
60  
PERC 3/SC Features  
SCSI Firmware  
The PERC 3/SC firmware handles all RAID and SCSI command processing  
and also supports the features shown in Table 4-8.  
Table 4-8. SCSI Firmware Support  
Feature  
Description  
Disconnect/  
reconnect  
Optimizes SCSI bus utilization  
Tagged command  
queuing  
Multiple tags to improve random access  
Scatter/gather  
Multi-threading  
Stripe size  
Single command can transfer data to and from different  
memory locations  
Up to 255 simultaneous commands with elevator sorting  
and concatenation of requests per SCSI channel  
Variable for all logical drives: 2 KB, 4 KB, 8 KB, 16 KB,  
32 KB, 64 KB, or 128 KB.  
Rebuild  
Multiple rebuilds and consistency checks with user-  
definable priority  
RAID Management  
RAID management is provided by software utilities that manage and  
configure the RAID system and PERC 3/SC, create and manage multiple  
disk arrays, control and monitor multiple RAID servers, provide error  
statistics logging, and provide online maintenance. They include:  
•
•
•
•
PERC 3 BIOS Configuration utility (Ctrl-M)  
Dell Manager  
WebBIOS Configuration Utility  
Dell™ OpenManage™ Array Manager  
PERC 3 BIOS Configuration Utility  
You can use the PERC 3 BIOS Configuration Utility to configure and  
maintain RAID arrays, format disk drives, and manage the RAID system. It  
is independent of any operating system. See Chapter 11 "PERC 3 BIOS  
Configuration Utility" for additional information.  
PERC 3/SC Features  
61  
Dell Manager  
Dell Manager is a character-based, non-GUI utility that changes policies  
and parameters, and monitors RAID systems. Dell Manager runs under  
Novell NetWare 5.x, 6.x, and Red Hat Linux 7.x. See Chapter 12 "Dell  
Manager" for additional information.  
WebBIOS Configuration Utility  
WebBIOS is used to configure and manage a RAID system using an HTML  
interface. See the WebBIOS Configuration Utility Guide for more  
information.  
Dell OpenManage Array Manager  
Dell OpenManage Array Manager is used to configure and manage a storage  
system that is connected to a server, while the server is active and continues  
to handle requests. Array Manager runs under NetWare, Windows NT, and  
Windows 2000. Refer to Dell documentation on the Dell web site at  
support.dell.com for more information.  
62  
PERC 3/SC Features  
SE CT ION 5  
PERC 3/DC and PERC  
3/DCL Features  
Hardware Requirements  
Configuration Features  
Hardware Architecture Features  
Array Performance Features  
Fault Tolerance Features  
Software Utilities  
Operating System Software Drivers  
PERC 3/DC and PERC 3/DCL Specifications  
Hardware Requirements  
NOTE: The PERC 3/DC  
has a battery option,  
while the PERC 3/DCL  
does not have a battery  
option.  
PERC 3/DC and PERC 3/DCL have two SCSI channels that support 160M  
and Wide SCSI at data transfer rates up to 160 MB/s. The SCSI channel  
supports up to 15 Wide devices or up to seven narrow devices.  
PERC 3/DC and PERC 3/DCL can be installed in a system with a  
motherboard that has 5 V, 32- or 64-bit PCI expansion slots. The system  
should have an Intel® Pentium® II or higher central processing unit (CPU),  
a diskette drive, a color monitor, video graphics adapter (VGA) card, and a  
keyboard. A mouse is recommended.  
Configuration Features  
The PERC 3/DC and PERC 3/DCL configuration features are shown in  
Table 5-1.  
Table 5-1. Configuration Features  
Specification  
Feature  
RAID levels  
0, 1, 5, 10, and 50  
SCSI channels  
2
Maximum number of drives per channel  
Array interface to host  
Drive interface  
15  
PCI 2.2  
Wide 160M  
Cache memory size  
64 MB (PERC 3/DCL)  
128 MB DIMM (PERC 3/DC)  
Cache function  
Write-back, write-through, adaptive  
read-ahead, non read-ahead, read-  
ahead  
Multiple logical drives/arrays per  
controller  
Up to 40 logical drives per controller  
Maximum number of PERC 3/DC and  
PERC 3/DCL controllers per system  
Contact your Dell™ representative.  
Online capacity expansion  
Yes  
Yes  
Dedicated and pool hot spare  
64  
PERC 3/DC and PERC 3/DCL Features  
Table 5-1. Configuration Features (continued)  
Specification  
Feature  
Flashable firmware  
Yes  
Hot swap devices supported  
Non-disk devices supported  
Mixed capacity hard disk drives  
Number of 16-bit internal connectors  
Cluster support  
Yes  
No  
Yes  
2
Yes (PERC 3/DC only)  
SMART Technology  
The PERC 3/DC and PERC 3/DCL self-monitoring analysis and reporting  
technology (SMART) detects predictable drive failures. SMART monitors  
the internal performance of all motors, heads, and drive electronics.  
Configuration on Disk  
Configuration on Disk (drive roaming) saves configuration information  
both in NVRAM on PERC 3/DC and PERC 3/DCL and on the disk drives  
attached to PERC 3/DC and PERC 3/DCL. If PERC 3/DC and PERC  
3/DCL are replaced, the new PERC 3/DC and PERC 3/DCL controllers  
detect the actual RAID configuration, maintaining the integrity of the data  
on each drive, even if the drives have changed channel and/or target ID.  
NOTE: Configuration on  
Disk does not work if you  
change both the adapter,  
and the SCSI connectors  
to different connectors on  
the new adapter. It works  
only if you make one  
change at a time.  
Drive roaming is supported across channels on the same controller. Drive  
roaming across channels is not supported when cluster mode is enabled.  
Table 5-2 shows the features for Configuration on Disk.  
Table 5-2. Configuration on Disk Features  
Specification  
Feature  
Support for hard disk drives with capacities of more  
than 8 gigabytes (GB)  
Yes  
Online RAID level migration  
RAID remapping  
Yes  
Yes  
No reboot necessary after expansion  
Yes (Only if virtual sizing  
is enabled.)  
PERC 3/DC and PERC 3/DCL Features  
65  
Table 5-2. Configuration on Disk Features (continued)  
Specification  
Feature  
Yes  
More than 200 qtags per array  
Hardware clustering support on the board  
User-specified rebuild rate  
Yes  
Yes  
Hardware Architecture Features  
Table 5-3 displays the PERC 3/DC and PERC 3/DCL hardware architecture  
features.  
Table 5-3. Hardware Architecture Features  
Specification  
Processor  
Feature  
Intel i960RN 100MHz  
®
SCSI controller(s)  
One QLogic 12160 Dual SCSI  
controller  
Size of flash ROM  
1 MB  
Amount of non-volatile random access  
memory (NVRAM)  
32 KB  
Hardware-exclusive OR (XOR)  
assistance  
Yes  
Yes  
Direct input/output (I/O)  
SCSI bus termination  
Active, single-ended or low voltage  
differential (LVD)  
Double-sided dual in-line memory  
modules (DIMMs)  
Yes  
Auxiliary TermPWR source  
Direct I/O bandwidth  
No  
533 MB/s (greater than 200 MB/s  
sustained)  
66  
PERC 3/DC and PERC 3/DCL Features  
Array Performance Features  
Table 5-4 displays the PERC 3/DC and PERC 3/DCL array performance  
features.  
Table 5-4. Array Per formance Features  
Specification  
Feature  
Host data transfer rate  
Drive data transfer rate  
Maximum scatter/gathers  
Maximum size of I/O requests  
Maximum queue tags per drive  
Stripe sizes  
533 MB/s  
160 MB/s  
26 elements  
6.4 MB in 64 KB stripes  
As many as the drive can accept.  
2 KB, 4 KB, 8 KB, 16 KB, 32 KB,  
64 KB, or 128 KB  
Maximum number of concurrent  
commands  
255  
Support for multiple initiations  
Yes  
Fault Tolerance Features  
Table 5-5 lists the fault tolerance features.  
Table 5-5. Fault Tolerance Features  
Specification  
Feature  
Support for SMART  
Yes  
Optional battery backup for cache memory  
Yes. Up to 72 hours data  
retention.  
Drive failure detection  
Automatic  
Automatic  
Yes  
Drive rebuild using hot spares  
Manual replacement of hard drives without system  
shutdown  
Parity generation and checking  
Yes  
PERC 3/DC and PERC 3/DCL Features  
67  
Software Utilities  
Table 5-6 lists the software utilities.  
Table 5-6. Software Utilities  
Specification  
Feature  
Yes  
Graphical user interface (GUI)  
Management utility  
Yes  
Bootup configuration using BIOS Configuration Utility  
(Ctrl–M)  
Yes  
Online read, write, and cache policy switching  
Intranet support  
Yes  
Yes  
Operating System Software Drivers  
PERC 3/DC and PERC 3/DCL include a Disk Operating System (DOS)  
software configuration utility and drivers for Microsoft® Windows NT®  
V4.0, Windows® 2000, Novell® NetWare® 5.x, 6.x, and Red Hat Linux 7.x.  
PERC 3/DC and PERC 3/DCL  
Specifications  
Table 5-7 lists the PERC 3/DC and PERC 3/DCL specifications.  
Table 5-7. PERC 3/DC and PERC 3/DCL Specifications  
Parameter  
Card size  
Specification  
Half-length PCI card size (6.875" X 4.2").  
Intel i960RN™ 64-bit RISC processor @ 100 MHz  
PCI 2.2  
Processor  
Bus type  
PCI controller  
Bus data transfer rate  
Intel i960RN  
Up to 533 MB/s at 64/66 MHz  
PERC 3/DC and PERC 3/DCL BIOS  
Basic input/output system  
(BIOS)  
68  
PERC 3/DC and PERC 3/DCL Features  
Table 5-7. PERC 3/DC and PERC 3/DCL Specifications  
Parameter  
Specification  
Cache configuration  
Firmware  
64 MB or 128 MB DIMM  
1 MB × 8 flash ROM  
Nonvolatile random access  
memory (RAM)  
32 KB × 8 for storing RAID configuration  
Operating voltage  
SCSI controller  
5 V, 3.3 V  
Two SCSI controllers for 160M and Wide support.  
Up to 160 MB/s per channel  
LVD or single-ended  
SCSI data transfer rate  
SCSI bus  
SCSI termination  
Termination disable  
Devices per SCSI channel  
SCSI device types  
RAID levels supported  
SCSI connectors  
Active  
Automatic through cable and device detection  
Up to 15 wide or seven narrow SCSI devices.  
Synchronous or asynchronous.  
0, 1, 5, 10, and 50  
Two 68-pin internal high-density connectors for 16-  
bit SCSI devices. Two ultra-high density 68-pin  
external connectors for 160M and Wide SCSI.  
Serial port  
3-pin RS232C-compatible connector (for  
manufacturing use only)  
PCI Bridge/CPU  
Both PERC 3/DC and PERC 3/DCL use the Intel i960RN PCI bridge with  
an embedded i960RN RISC processor running at 100 MHz. The RN bridge  
handles data transfers between the primary (host) PCI bus, the secondary  
PCI bus, cache memory, and the SCSI bus. The DMA controller supports  
chaining and unaligned data transfers. The embedded i960JN CPU directs  
all controller functions, including command processing, SCSI bus transfers,  
RAID processing, drive rebuilding, cache management, and error recovery.  
PERC 3/DC and PERC 3/DCL Features  
69  
Cache Memory  
Either 64 MB or 128 MB of PERC 3/DC and PERC 3/DCL cache memory  
reside in a memory bank that uses a 64 MB or 128 MB SDRAM DIMM.  
PERC 3/DC and PERC 3/DCL support write-through or write-back  
caching, selectable for each logical drive. To improve performance in  
sequential disk accesses, the PERC 3/DC and PERC 3/DCL controllers use  
read-ahead caching by default. You can disable read-ahead caching.  
PERC 3/DC and PERC 3/DCL BIOS  
The BIOS resides on a 1 MB × 8 flash ROM for easy upgrade. The PERC  
3/DC and PERC 3/DCL BIOS supports INT 13h calls to boot DOS without  
special software or device drivers. The PERC 3/DC and PERC 3/DCL BIOS  
provides an extensive setup utility that can be accessed by pressing <Ctrl>  
<M> at BIOS initialization. PERC 3/DC and PERC 3/DCL BIOS  
Configuration utility is described in Chapter 11 "PERC 3 BIOS  
Configuration Utility."  
Onboard Speaker  
The PERC 3/DC and PERC 3/DCL controller has an onboard tone  
generator for audible warnings when system errors occur. Audible warnings  
can be generated through this speaker. The audible warnings are listed in  
Chapter 13 "Troubleshooting."  
Serial Port  
PERC 3/DC and PERC 3/DCL includes a 3-pin RS232C-compatible serial  
port connector. This is for manufacturing use only.  
SCSI Bus  
The PERC 3/DC and PERC 3/DCL controller has four Wide 160M SCSI  
channels that support both LVD and single-ended devices with active  
termination. Synchronous and asynchronous devices are supported. PERC  
3/DC and PERC 3/DCL provides automatic termination disable via cable  
detection. The SCSI channel supports up to 15 wide or seven non-wide  
SCSI devices at speeds up to 160 MB/s.  
70  
PERC 3/DC and PERC 3/DCL Features  
SCSI Connectors  
The PERC 3/DC and PERC 3/DCL adapter has two types of SCSI  
connectors:  
•
•
Two 68-pin high density internal connectors  
Two 68-pin external ultra-high-density connectors  
Only one connector type can be used for the SCSI channel at any one time.  
SCSI Termination  
PERC 3/DC and PERC 3/DCL uses active termination on the SCSI bus  
conforming to SCSI-2 and SCSI-3 specifications. Termination  
enable/disable is automatic through cable detection.  
SCSI Firmware  
The PERC 3/DC and PERC 3/DCL firmware handles all RAID and SCSI  
command processing and supports the feature listed in Table 5-8.  
Table 5-8. SCSI Firmware  
Feature  
Description  
Disconnect/  
reconnect  
Optimizes SCSI bus seek.  
Tagged command  
queuing  
Multiple tags to improve random access  
Scatter/gather  
Multi-threading  
Stripe size  
A single command can transfer data to and from different  
memory locations.  
Up to 255 simultaneous commands with elevator sorting  
and concatenation of requests per SCSI channel  
Variable for all logical drives: 2 KB, 4 KB, 8 KB, 16 KB,  
32 KB, 64 KB, or 128 KB  
Rebuild  
Multiple rebuilds and consistency checks with user-  
definable priority  
PERC 3/DC and PERC 3/DCL Features  
71  
RAID Management  
RAID management is provided by software utilities that manage and  
configure the RAID system and PERC 3/DC and PERC 3/DCL, create and  
manage multiple disk arrays, control and monitor multiple RAID servers,  
provide error statistics logging, and provide online maintenance. The  
utilities are:  
•
•
•
•
PERC 3 BIOS Configuration utility (Ctrl-M)  
Dell Manager  
WebBIOS Configuration Utility  
Dell OpenManage™ Array Manager  
PERC 3 BIOS Configuration Utility  
You can use the PERC 3 BIOS Configuration Utility to configure and  
maintain RAID arrays, format disk drives, and manage the RAID system. It  
is independent of any operating system. See Chapter 11 "PERC 3 BIOS  
Configuration Utility" for additional information.  
Dell Manager  
Dell Manager is a character-based, non-GUI utility that changes policies,  
and parameters, and monitors RAID systems. Dell Manager runs under  
Novell NetWare 5.x, 6.x, and Red Hat Linux 7.x. See Chapter 12 "Dell  
Manager" for additional information.  
WebBIOS Configuration Utility  
WebBIOS is used to configure and manage a RAID system using an HTML  
interface. See the WebBIOS Configuration Utility Guide for more  
information.  
Dell OpenManage Array Manager  
Dell OpenManage Array Manager is used to configure and manage a storage  
system that is connected to a server, while the server is active and continues  
to handle requests. Array Manager runs under NetWare, Windows NT, and  
Windows 2000. Refer to Dell documentation on the Dell web site at  
support.dell.com for more information.  
72  
PERC 3/DC and PERC 3/DCL Features  
SE CT ION 6  
PERC 3/QC Features  
Hardware Requirements  
Configuration Features  
Hardware Architecture Features  
Array Performance Features  
Fault Tolerance Features  
Software Utilities  
Operating System Software Drivers  
PERC 3/QC Specifications  
Hardware Requirements  
PERC 3/QC is a high-performance intelligent peripheral component  
interconnect (PCI)-to-SCSI host adapter with RAID control capabilities.  
PERC 3/QC has four SCSI channels that support 160M and Wide SCSI at  
data transfer rates up to 160 MB/s. Each SCSI channel supports up to 15  
Wide devices or up to seven narrow devices.  
PERC 3/QC can be installed in a system with a motherboard that has 5 V,  
32- or 64-bit PCI expansion slots. The system should have an Intel®  
Pentium® II or higher central processing unit (CPU), a diskette drive, a  
color monitor, video graphics adapter (VGA) card, and a keyboard. A mouse  
is recommended.  
Configuration Features  
Table 6-1 lists the PERC 3/QC configuration features.  
Table 6-1. Configuration Features  
Specification  
Feature  
0, 1, 5, 10, and 50  
4
RAID levels  
SCSI channels  
Maximum number of drives per channel  
Array interface to host  
Drive interface  
15  
PCI 2.2  
Wide 160M  
128 MB DIMM  
Cache memory size  
Cache function  
Write-back, write-through, adaptive  
read-ahead, non read-ahead, read-  
ahead  
Multiple logical drives/arrays per  
controller  
Up to 40 logical drives per controller  
Maximum number of PERC 3/QC  
controllers per system  
Contact your Dell™ representative.  
Online capacity expansion  
Dedicated and pool hot spare  
Flashable firmware  
Yes  
Yes  
Yes  
74  
PERC 3/QC Features  
Table 6-1. Configuration Features (continued)  
Specification  
Feature  
Yes  
Hot swap devices supported  
Non-disk devices supported  
Mixed-capacity hard drives  
Number of 16-bit internal connectors  
No  
Yes  
2
SMART Technology  
NOTE: The PERC 3/QC  
does not support  
clustering.  
The PERC 3/QC self-monitoring analysis and reporting technology  
(SMART) detects predictable drive failures. SMART monitors the internal  
performance of all motors, heads, and drive electronics.  
Configuration on Disk  
Configuration on Disk (drive roaming) saves configuration information  
both in non-volatile random access memory (NVRAM) on PERC 3/QC and  
on the disk drives attached to PERC 3/QC. If PERC 3/QC is replaced, the  
new PERC 3/QC controller detects the actual RAID configuration,  
maintaining the integrity of the data on each drive, even if the drives have  
changed channel and/or target ID.  
NOTE: Configuration on  
Disk does not work if you  
change both the adapter,  
and the SCSI connectors  
to different connectors on  
the new adapter. It works  
only if you make one  
change at a time.  
Drive roaming is supported across channels on the same controller. Drive  
roaming across channels is not supported when cluster mode is enabled.  
Table 6-2 lists the Configuration on Disk features.  
Table 6-2. Configuration on Disk Features  
Specification  
Feature  
Support for hard drives with capacities of more  
than 8 gigabytes (GB)  
Yes  
Online RAID level migration  
RAID remapping  
Yes  
Yes  
No reboot necessary after expansion  
Yes (only if virtual sizing is  
enabled)  
More than 200 qtags per array  
Yes  
Yes  
Hardware clustering support on the board  
PERC 3/QC Features  
75  
Table 6-2. Configuration on Disk Features (continued)  
Specification  
Feature  
User-specified rebuild rate  
Yes  
Hardware Architecture Features  
Table 6-3 lists the PERC 3/QC hardware architecture features.  
Table 6-3. Hardware Architecture Features  
Specification  
Processor  
Feature  
Intel i960RN 100MHz  
®
SCSI controller(s)  
Two QLogic 12160 dual SCSI  
controller  
Size of flash ROM  
Amount of NVRAM  
1 MB  
32 KB  
Yes  
Hardware exclusive OR (XOR)  
assistance  
Direct input/output (I/O)  
SCSI bus termination  
Yes  
Active, single-ended or low-voltage  
differential (LVD)  
Double-sided dual in-line memory  
modules (DIMMs)  
Yes  
Auxiliary TermPWR source  
Direct I/O bandwidth  
No  
533 MB/s (greater than 200 MB/s  
sustained)  
Array Performance Features  
Table 6-4 lists the array performance features.  
Table 6-4. Array Per formance Features  
Specification  
Feature  
Host data transfer rate  
533 MB/s  
76  
PERC 3/QC Features  
Table 6-4. Array Per formance Features (continued)  
Specification  
Feature  
Drive data transfer rate  
Maximum scatter/gathers  
Maximum size of I/O requests  
Maximum queue tags per drive  
Stripe sizes  
160 MB/s  
26 elements  
6.4 MB in 64 KB stripes  
As many as the drive can accept.  
2 KB, 4 KB, 8 KB, 16 KB, 32 KB,  
64 KB, or 128 KB  
Maximum number of concurrent  
commands  
255  
Support for multiple initiations  
Yes  
Fault Tolerance Features  
The fault tolerance features are listed in Table 6-5.  
Table 6-5. Fault Tolerance Features  
Specification  
Feature  
Support for SMART  
Yes  
Optional battery backup for cache memory  
Yes. Up to 72 hours data  
retention.  
Drive failure detection  
Automatic  
Automatic  
Yes  
Drive rebuild using hot spares  
Manual replacement of drives without system  
shutdown  
Parity generation and checking  
Yes  
Software Utilities  
The software utilities are listed in Table 6-6.  
PERC 3/QC Features  
77  
Table 6-6. Software Utilities  
Specification  
Feature  
Yes  
Graphical user interface  
Management utility  
Yes  
Bootup configuration using PERC BIOS configuration  
utility (Ctrl–M)  
Yes  
Online read, write, and cache policy switching  
Intranet support  
Yes  
Yes  
Operating System Software Drivers  
PERC 3/QC includes a DOS software configuration utility and drivers for  
Microsoft® Windows NT® 4.0, Windows® 2000, Novell® NetWare® and  
Red Hat Linux.  
PERC 3/QC Specifications  
The PERC 3/QC specifications are listed in Table 6-7.  
Table 6-7. PERC 3/QC Specifications  
Parameter  
Card size  
Processor  
Specification  
12.3" x 4.2" (full length PCI)  
Intel i960RN™ 64-bit RISC processor @ 100  
MHz  
Bus type  
PCI 2.2  
PCI controller  
Bus data transfer rate  
Intel i960RN  
Up to 533 MB/s at 64/66 MHz  
PERC 3/QC BIOS  
Basic input/output system  
(BIOS)  
Cache configuration  
Firmware  
64 MB or 128 MB DIMM  
1 MB × 8 flash ROM  
Non-volatile random access  
memory (RAM)  
32 KB × 8 for storing RAID configuration  
78  
PERC 3/QC Features  
Table 6-7. PERC 3/QC Specifications (continued)  
Parameter  
Specification  
Operating voltage  
SCSI controller  
5 V, 3.3 V  
Four SCSI controllers for 160M and Wide  
support.  
SCSI data transfer rate  
SCSI bus  
Up to 160 MB/s per channel  
LVD or single-ended  
SCSI termination  
Termination disable  
Devices per SCSI channel  
SCSI device types  
RAID levels supported  
SCSI connectors  
Active  
Automatic through cable and device detection  
Up to 15 wide or seven narrow SCSI devices.  
Synchronous or asynchronous.  
0, 1, 5, 10, and 50  
Two 68-pin internal high-density connectors for  
16-bit SCSI devices. Four ultra-high density 68-  
pin external connectors for 160M and Wide  
SCSI.  
Serial port  
9-pin RS232C-compatible connector (for  
manufacturing use only)  
PCI Bridge/CPU  
PERC 3/QC uses the Intel i960RN PCI bridge with an embedded i960RN  
RISC processor running at 100 MHz. The RN bridge handles data transfers  
between the primary (host) PCI bus, the secondary PCI bus, cache memory,  
and the SCSI bus. The DMA controller supports chaining and unaligned  
data transfers. The embedded i960JN CPU directs all controller functions,  
including command processing, SCSI bus transfers, RAID processing, drive  
rebuilding, cache management, and error recovery.  
Cache Memory  
128 MB of PERC 3/QC cache memory resides in a memory bank that uses a  
64 MB or 128 MB SDRAM DIMM. PERC 3/QC supports write-through or  
write-back caching, selectable for each logical drive. To improve  
performance in sequential disk accesses, the PERC 3/QC controller uses  
read-ahead caching by default. You can disable read-ahead caching.  
PERC 3/QC Features  
79  
PERC 3/QC BIOS  
The BIOS resides on a 1 MB × 8 flash ROM for easy upgrade. The PERC  
3/QC BIOS supports INT 13h calls to boot DOS without special software or  
device drivers. The PERC 3/QC BIOS provides an extensive setup utility  
that can be accessed by pressing <Ctrl> <M> at BIOS initialization.  
PERC 3 BIOS Configuration Utility is described in Chapter 11  
"Troubleshooting."  
Onboard Speaker  
The PERC 3/QC controller has an onboard tone generator for audible  
warnings when system errors occur. Audible warnings can be generated  
through this speaker. The audible warnings are listed in Chapter 13.  
Serial Port  
PERC 3/QC includes a 9-pin RS232C-compatible serial port connector.  
This is for manufacturing use only.  
SCSI Bus  
PERC 3/QC has four Wide 160M SCSI channels that support both LVD  
and single-ended devices with active termination. Synchronous and  
asynchronous devices are supported. PERC 3/QC provides automatic  
termination disable by using cable detection. The SCSI channel supports  
up to 15 wide or seven non-wide SCSI devices at speeds up to 160 MB/s.  
SCSI Connectors  
PERC 3/QC has two types of SCSI connectors:  
•
•
Two 68-pin high density internal connectors  
Four 68-pin external ultra-high-density connectors  
Only one connector type can be used for the SCSI channel at any one time.  
SCSI Termination  
PERC 3/QC uses active termination on the SCSI bus conforming to SCSI-2  
and SCSI-3 specifications. Termination enable/disable is automatic through  
cable detection.  
80  
PERC 3/QC Features  
SCSI Firmware  
The PERC 3/QC firmware handles all RAID and SCSI command  
processing and supports the features described in Table 6-8.  
Table 6-8. SCSI Firmware  
Feature  
Description  
Disconnect/  
reconnect  
Optimizes SCSI bus utilization  
Tagged command  
queuing  
Multiple tags to improve random access  
Scatter/gather  
Multi-threading  
Stripe size  
A single command can transfer data to and from different  
memory locations.  
Up to 255 simultaneous commands with elevator sorting  
and concatenation of requests per SCSI channel  
Variable for all logical drives: 2 KB, 4 KB, 8 KB, 16 KB,  
32 KB, 64 KB, or 128 KB.  
Rebuild  
Multiple rebuilds and consistency checks with user-  
definable priority  
RAID Management  
RAID management is provided by software utilities that manage and  
configure the RAID system and PERC 3/QC, create and manage multiple  
disk arrays, control and monitor multiple RAID servers, provide error  
statistics logging, and provide online maintenance. They include:  
•
•
•
•
PERC 3/QC BIOS Configuration Utility (Ctrl-M)  
Dell Manager  
WebBIOS Configuration Utility  
Dell™ OpenManage™ Array Manager  
PERC 3 BIOS Configuration Utility  
You can use the PERC 3 BIOS Configuration Utility to configure and  
maintain RAID arrays, format disk drives, and manage the RAID system. It  
is independent of any operating system. See Chapter 11 "PERC 3 BIOS  
Configuration Utility" for additional information.  
PERC 3/QC Features  
81  
Dell Manager  
Dell Manager is a character-based, non-GUI utility that changes policies,  
and parameters, and monitors RAID systems. Dell Manager runs under  
Novell NetWare 5.x, 6.x, and Red Hat Linux 7.x. See Chapter 12 "Dell  
Manager" for additional information.  
WebBIOS Configuration Utility  
WebBIOS is used to configure and manage a RAID system using an HTML  
interface. See the WebBIOS Configuration Utility Guide for more  
information.  
Dell OpenManage™ Array Manager  
Dell OpenManage Array Manager is used to configure and manage a storage  
system that is connected to a server, while the server is active and continues  
to handle requests. Array Manager runs under NetWare, Windows NT, and  
Windows 2000. Refer to Dell documentation on the Dell web site at  
support.dell.com for more information.  
82  
PERC 3/QC Features  
SE CT ION 7  
Configuring PERC 3  
Configuring SCSI Physical Drives  
Current Configuration  
Logical Drive Configuration  
Physical Device Layout  
Configuring Arrays  
Configuration Strategies  
Assigning RAID Levels  
Configuring Logical Drives  
Optimizing Data Storage  
Planning the Array Configuration  
Random Array Deletion  
Configuring SCSI Physical Drives  
Physical SCSI drives must be organized into logical drives. The arrays and  
logical drives that you construct must be able to support the RAID) level  
that you select.  
Observe the following guidelines when connecting and configuring SCSI  
devices in a RAID array:  
•
•
You can place up to 32 physical disk drives in an array.  
Include all drives that have the same capacity in the same array. If the  
drives are not the same size, the array uses the size of the smallest drive  
and the same amount of space on the other drives to construct the  
arrays.  
•
•
Make sure any hot spare has a capacity that is at least as large as the  
largest drive that may be replaced by the hot spare.  
When replacing a failed drive, make sure that the replacement drive  
has a capacity that is at least as large as the drive being replaced.  
Current Configuration  
Your PERC 3 card has one (SC), two (DC, DCL), or four channels (QC),  
depending on which card you have. Use Table 7-1 to list the devices that  
you assign to each SCSI ID for SCSI Channel 0.  
Table 7-1. Current Configuration for SCSI  
Channel 0  
SCSI ID  
Device Description  
SCSI Channel 0  
0
1
2
3
4
5
6
84  
Configuring PERC 3  
Table 7-1. Current Configuration for SCSI  
Channel 0 (continued)  
SCSI ID  
Device Description  
SCSI Channel 0  
8
9
10  
11  
12  
13  
14  
15  
Use Table 7-2 to list the devices that you assign to each SCSI ID for SCSI  
Channel 1.  
Table 7-2. Current Configuration for SCSI  
Channel 1  
SCSI ID  
Device Description  
SCSI Channel 1  
0
1
2
3
4
5
6
8
9
10  
11  
12  
Configuring PERC 3  
85  
Table 7-2. Current Configuration for SCSI  
Channel 1 (continued)  
SCSI ID  
Device Description  
SCSI Channel 1  
13  
14  
15  
Use Table 7-3 to list the devices that you assign to each SCSI ID for SCSI  
Channel 2.  
Table 7-3. Current Configuration for SCSI  
Channel 2  
SCSI ID  
Device Description  
SCSI Channel 2  
0
1
2
3
4
5
6
8
9
10  
11  
12  
13  
14  
15  
86  
Configuring PERC 3  
Use Table 7-4 to list the devices that you assign to each SCSI ID for SCSI  
Channel 3.  
Table 7-4. Current Configuration for SCSI  
Channel 3  
SCSI ID  
Device Description  
SCSI Channel 3  
0
1
2
3
4
5
6
8
9
10  
11  
12  
13  
14  
15  
Logical Drive Configuration  
Use Table 7-5 to list the details for each logical drive that you configure.  
Table 7-5. Logical Drive Configuration  
Logical  
Drive  
RAID Stripe  
Size  
Logical  
Drive Size  
Cache  
Policy  
Read  
Policy  
Write  
Policy  
# of Physical  
Drives  
LD0  
LD1  
Configuring PERC 3  
87  
Logical  
Drive  
RAID Stripe  
Size  
Logical  
Drive Size  
Cache  
Policy  
Read  
Policy  
Write  
Policy  
# of Physical  
Drives  
LD2  
LD3  
LD4  
LD5  
LD6  
LD7  
LD8  
LD9  
LD10  
LD11  
LD12  
LD13  
LD14  
LD15  
LD16  
LD17  
LD18  
LD19  
LD20  
LD21  
LD22  
LD23  
LD24  
LD25  
LD26  
LD27  
LD28  
LD29  
LD30  
88  
Configuring PERC 3  
Logical  
Drive  
RAID Stripe  
Size  
Logical  
Drive Size  
Cache  
Policy  
Read  
Policy  
Write  
Policy  
# of Physical  
Drives  
LD31  
LD32  
LD33  
LD34  
LD35  
LD36  
LD37  
LD38  
LD39  
Physical Device Layout  
Use Table 7-6 to list the details for each physical device on the channels.  
Table 7-6. Physical Device L ayout  
Channel 0  
Channel 1  
Channel 2  
Channel 3  
Target ID  
Device type  
Logical drive number/  
Drive number  
Manufacturer/Model  
number  
Firmware level  
Target ID  
Device type  
Logical drive number/  
Drive number  
Manufacturer/Model  
number  
Firmware level  
Target ID  
Configuring PERC 3  
89  
Table 7-6. Physical Device L ayout (continued)  
Channel 0 Channel 1  
Channel 2  
Channel 3  
Device type  
Logical drive number/  
Drive number  
Manufacturer/Model  
number  
Firmware level  
Target ID  
Device type  
Logical drive number/  
Drive number  
Manufacturer/Model  
number  
Firmware level  
Target ID  
Device type  
Logical drive number/  
Drive number  
Manufacturer/Model  
number  
Firmware level  
Target ID  
Device type  
Logical drive number/  
Drive number  
Manufacturer/Model  
number  
Firmware level  
Target ID  
Device type  
Logical drive number/  
Drive number  
90  
Configuring PERC 3  
Table 7-6. Physical Device L ayout (continued)  
Channel 0 Channel 1  
Channel 2  
Channel 3  
Manufacturer/Model  
number  
Firmware level  
Target ID  
Device type  
Logical drive number/  
Drive number  
Manufacturer/Model  
number  
Firmware level  
Target ID  
Device type  
Logical drive number/  
Drive number  
Manufacturer/Model  
number  
Firmware level  
Target ID  
Device type  
Logical drive number/  
Drive number  
Manufacturer/Model  
number  
Firmware level  
Target ID  
Device type  
Logical drive number/  
Drive number  
Manufacturer/Model  
number  
Firmware level  
Target ID  
Configuring PERC 3  
91  
Table 7-6. Physical Device L ayout (continued)  
Channel 0 Channel 1  
Channel 2  
Channel 3  
Device type  
Logical drive number/  
Drive number  
Manufacturer/Model  
number  
Firmware level  
Target ID  
Device type  
Logical drive number/  
Drive number  
Manufacturer/Model  
number  
Firmware level  
Target ID  
Device type  
Logical drive number/  
Drive number  
Manufacturer/Model  
number  
Firmware level  
Configuring Arrays  
Connect the physical drives to the RAID controller, configure the disks,  
then initialize them. PERC 3 supports up to 40 arrays. An array is a group of  
physical disk drives. An array can consist of one to 32 physical disk drives  
(256 drives when used with the span feature in a RAID 50 configuration).  
The number of drives in an array determines the RAID levels that can be  
supported. The maximum number of drives that can be used is 60,  
consisting of four channels with 15 drives each. PERC 3 supports up to 40  
logical drives per controller.  
92  
Configuring PERC 3  
Arranging Arrays  
You must arrange the arrays to provide additional organization for the drive  
array. You must arrange arrays so that you can create system drives that can  
function as boot devices.  
You can sequentially arrange arrays with an identical number of drives so  
that the drives in the group are spanned. Spanned drives can be treated as  
one large drive. Data can be striped across multiple arrays as one logical  
drive.  
You can create spanned drives using the PERC 3 BIOS Configuration  
Utility.  
Creating Hot Spares  
Any drive that is present, formatted, and initialized but is not included in  
an array or logical drive can be designated as a hot spare.  
You can designate drives as hot spares using the PERC 3 BIOS  
Configuration Utility.  
Creating Logical Drives  
Logical drives are arrays or spanned arrays that are presented to the  
operating system. You must create one or more logical drives. The logical  
drive capacity must include all of the disk space in an array. If an array with  
drives with mixed sizes, the smallest common size is used and larger disk  
drives are truncated. The logical drive capacity can also be larger than an  
array by using spanning. PERC 3 supports up to 40 logical drives.  
Configuration Strategies  
The most important factors in RAID array configuration are:  
•
•
•
Drive capacity  
Drive availability (fault tolerance)  
Drive performance  
You cannot configure a logical drive that optimizes all three factors, but it is  
easy to choose a logical drive configuration that maximizes one factor at the  
expense of the other two.  
Configuring PERC 3  
93  
Maximizing Capacity  
RAID 0 achieves maximum drive capacity, but does not provide data  
redundancy. Maximum drive capacity for each RAID level is shown below.  
original equipment manufacturer (OEM) level firmware that can span up to  
four logical drives is assumed. Table 7-7 displays the drives required, and  
capacity for the various RAID levels.  
Table 7-7. Drives and Capacities for Each RAID Level  
RAID  
Level  
Description  
Drives  
Required  
Capacity  
0
Striping without  
parity  
1 – 32  
(Number of disks) X (capacity of  
smallest disk)  
1
5
Mirroring  
2
(Capacity of smallest disk) X (1)  
Striping with  
floating parity  
drive  
3 – 32  
(Number of disks) X (capacity of  
smallest disk) - (capacity of 1  
disk)  
10  
50  
RAID 1 and  
striping  
4 – 16  
(Number of disks) X (capacity of  
smallest disk) / (2)  
(Must be a  
multiple of  
2)  
RAID 5 and  
striping  
6 – 256  
(Number of disks) X (capacity of  
smallest disk) – (capacity of 1  
disk X number of arrays)  
(Must be a  
multiple of  
arrays)  
Maximizing Drive Availability  
You can maximize the availability of data on the physical disk drive in the  
logical array by maximizing the level of fault tolerance. The levels of fault  
tolerance provided by the RAID levels are shown in Table 7-8.  
Table 7-8. Fault Tolerance Features for RAID Levels 0, 1  
and 5  
RAID  
Level  
Fault Tolerance Protection  
0
1
No fault tolerance  
Disk mirroring, which provides 100% data redundancy.  
94  
Configuring PERC 3  
Table 7-8. Fault Tolerance Features for RAID Levels 0, 1  
and 5 (continued)  
RAID  
Level  
Fault Tolerance Protection  
5
100% protection through striping and parity. The data is striped and  
parity data is written across a number of physical disk drives.  
Maximizing Drive Performance  
You can configure an array for optimal performance; however, optimal drive  
configuration for one type of application will probably not be optimal for  
any other application. A basic guideline of the performance characteristics  
for RAID drive arrays at each RAID level is shown in Table 7-9.  
Table 7-9. Performance Characteristics for Each RAID  
Level  
RAID  
Level  
Performance Characteristics  
0
Excellent for all types of input/output (I/O) activity, but provides no  
data security.  
1
5
Provides data redundancy and good performance.  
Provides data redundancy and good performance in most  
environments.  
10  
50  
Provides high data transfers and complete data redundancy.  
Provides high data transfers and data redundancy.  
Assigning RAID Levels  
Only one RAID level can be assigned to each logical drive. The drives  
required per RAID level shown in Table 7-10.  
Table 7-10. Physical Drives Required for Each RAID Level  
RAID  
Level  
Minimum Number of  
Physical Drives  
Maximum Number of Physical  
Drives  
0
1
1
2
32  
2
Configuring PERC 3  
95  
Table 7-10. Physical Drives Required for Each RAID Level  
RAID  
Level  
Minimum Number of  
Physical Drives  
Maximum Number of Physical  
Drives  
5
3
4
6
32  
16  
10  
50  
256  
(PERC has a limit of 60, using  
PERC 4/QC with 15 drives on each  
of the 4 channels.)  
Configuring Logical Drives  
After you have installed the PERC 3 controller in the server and have  
attached all physical disk drives, perform the following steps to prepare a  
RAID disk array:  
1
2
3
4
Optimize the PERC 3 controller options for your system.  
Press <Ctrl> <M> to run the BIOS Configuration Utility.  
Define and configure one or more logical drives.  
Select Easy Configuration in the PERC 3 BIOS Configuration Utility  
or select New Configuration to customize the RAID array.  
5
6
7
8
Create and configure one or more system drives (logical drives).  
Select the RAID level, cache policy, read policy, and write policy.  
Save the configuration.  
Initialize the system drives.  
After initialization, you can install the operating system.  
Optimizing Data Storage  
Data Access Requirements  
Each type of data stored in the disk subsystem has a different frequency of  
read and write activity. If you know the data access requirements, you can  
more successfully determine a strategy for optimizing the disk subsystem  
capacity, availability, and performance.  
96  
Configuring PERC 3  
Servers that support Video on Demand typically read the data often, but  
write data infrequently. Both the read and write operations tend to be long.  
Data stored on a general-purpose file server involves relatively short read  
and write operations with relatively small files.  
Array Functions  
You must first define the major purpose of the disk array. Will this disk array  
increase the system storage capacity for general-purpose file and print  
servers? Does this disk array support any software system that must be  
available 24 hours per day? Will the information stored in this disk array  
contains large audio or video files that must be available on demand? Will  
this disk array contain data from an imaging system?  
You must identify the purpose of the data to be stored in the disk subsystem  
before you can confidently choose a RAID level and a RAID configuration.  
Planning the Array Configuration  
Complete the items in Table 7-11 to help you plan the array configuration.  
Table 7-11. Factors to Consider for Array Configuration  
Question  
Answer  
Number of physical disk drives in the array  
Purpose of this array. Rank the following factors:  
Maximize drive capacity  
Maximize the safety of the data (fault tolerance)  
Maximize hard drive performance and throughput  
How many hot spares?  
Amount of cache memory installed on PERC 3  
Are all of the disk drives and the server protected by a UPS?  
Configuring PERC 3  
97  
Using the Array Configuration Planner  
Table 7-12 lists RAID levels, fault tolerance, and effective capacity for some  
possible drive configurations for an array consisting of one to 32 drives. This  
table does not take into account any hot spare (standby) drives. You should  
always have a hot spare drive in case of drive failure. RAID 1 requires two  
drives. RAID 5 requires at least three drives. RAID 10 requires at least four  
drives. RAID 50 requires at least six drives.  
Table 7-12. Array Configuration Planner  
Number  
of Drives  
Possible RAID  
Levels  
Relative  
Performance Tolerance  
Fault  
Effective  
Capacity  
1
2
2
3
3
4
4
4
5
5
6
6
6
RAID 0  
RAID 0  
RAID 1  
RAID 0  
RAID 5  
RAID 0  
RAID 5  
RAID 10  
RAID 0  
RAID 5  
RAID 0  
RAID 5  
Excellent  
Excellent  
Good  
No  
No  
Yes  
No  
Yes  
No  
Yes  
Yes  
No  
Yes  
No  
Yes  
Yes  
100%  
100%  
50%  
Excellent  
Good  
100%  
67%  
Excellent  
Good  
100%  
75%  
Good  
50%  
Excellent  
Good  
100%  
80%  
Excellent  
Good  
100%  
83%  
RAID 10  
(spanned)  
Excellent  
50%  
6
RAID 50  
(spanned)  
Good  
Yes  
67%  
7
7
8
8
8
RAID 0  
RAID 5  
RAID 0  
RAID 5  
Excellent  
Good  
No  
Yes  
No  
Yes  
Yes  
100%  
86%  
Excellent  
Good  
100%  
87%  
RAID 10  
(spanned)  
Excellent  
50%  
98  
Configuring PERC 3  
Table 7-12. Array Configuration Planner (continued)  
Number  
of Drives  
Possible RAID  
Levels  
Relative  
Performance Tolerance  
Fault  
Effective  
Capacity  
8
RAID 50  
(spanned)  
Good  
Yes  
Yes  
Yes  
75%  
50%  
88%  
16  
16  
RAID 10  
(spanned)  
Excellent  
Good  
RAID 50  
(spanned)  
32  
32  
32  
RAID 0  
Excellent  
Good  
No  
Yes  
Yes  
100%  
50%  
94%  
RAID 10  
RAID 50  
Good  
Random Array Deletion  
The PERC controllers support random array deletion, which is the ability to  
delete any unwanted logical drives and use that space for a new logical drive.  
This section describes the impact of random array deletion from the user  
perspective. It discusses the configuration module architecture  
modification, and the data structure.  
Overview  
NOTE: Refer to Chapter  
11 and Chapter 12 for the  
procedures for random  
array deletion.  
On a running operating system, you can not delete a physical drive entry in  
the drive object repository. Additions and deletions can be done only at the  
end of the repository. The operating system driver receives this support from  
the firmware and maintains the same object repository table for that session  
of the operating system.  
Configuring PERC 3  
99  
NOTE: When a ‘delete’  
request reaches the  
After you delete a logical drive, you can create a new one. You can use the  
configuration utilities to create the next logical drive from the non-  
contiguous free space (‘holes’), and from the newly created arrays. The  
configuration utility provides a list of configurable arrays where there is a  
space to configure.  
operating system driver,  
the driver stops all the  
running input/output (I/O)  
for other logical drives  
and processes the delete  
request first. Normal  
read/write operation  
starts after the delete  
request is completed.  
NOTICE: The deletion of the logical drive can fail under certain conditions.  
You cannot delete a logical drive during a reconstruction. Deletion can fail  
during a rebuild, initialization or check consistency of a logical drive, if that  
drive has a higher logical drive number than the drive you want to delete.  
Configuration Module  
The main benefit of random array deletion on the configuration module is  
that you are not restricted to sequential or contiguous logical drives when  
you create logical drives. You can use non-contiguous segments to create  
logical drives.  
NOTE: Drive size  
The random deletion of logical drives creates non-contiguous segments in  
the configuration, which prevents the sequential creation of the next logical  
drive. You can create logical drives from these non-contiguous segments. To  
create such a logical drive, you can span these segments, as long as they have  
the same number of physical drives.  
expansion is not possible,  
even though you can use  
non-contiguous free space  
to create a new logical  
drive.  
NOTE: You cannot move  
an existing logical drive  
to another area to protect  
it from defragmentation  
caused by random  
You can still create sequential logical drives, without using the non-  
contiguous segments. The utilities provide information about sequential  
segments, non-contiguous segments and physical drives that have not been  
configured. You can use this information when you create logical drives.  
deletion.  
100  
Configuring PERC 3  
SE CT ION 8  
PERC 3/SC Hardware  
Installation  
Requirements  
PERC 3/SC Card Layout  
Installation Steps  
Requirements  
This chapter describes the installation procedures for the PERC 3/SC  
board. You must have the following:  
•
•
•
•
A PERC 3/SC controller  
A host computer with an available 5 V, 32- or 64-bit PCI expansion slot  
The PERC 3 installation diskettes  
The necessary SCSI cables (This depends on the number and type of  
SCSI devices to be attached.)  
•
Fast, Ultra, Ultra II, and 160M SCSI hard drives  
Dell™ strongly recommends that you have an uninterruptible power supply  
(UPS) for the entire system.  
Optional Equipment  
You may also want to install SCSI cables that connect the PERC 3/SC to  
external SCSI devices.  
102  
PERC 3/SC Hardware Installation  
PERC 3/SC Card Layout  
Figure 8-1 shows the PERC 3/SC card, jumpers, and connectors.  
Figure 8-1. PERC 3/SC Card L ayout  
J1 SCSI Bus Termination  
Enable Control  
J11  
Internal  
Straddle Mount  
J4 Serial EPROM  
J5 Serial Port  
J10  
J2 CPLD  
SCSI Bus  
Termination Power  
Programming  
J6 Write Pending  
J7 Bios Enable  
J3 NVRAM Clear  
J8 User Activity  
LED  
J1  
J3 J4 J5 J6 J7 J8  
J9  
J2  
J12  
J9 I2C  
Connector  
J13  
External  
Connector for  
Battery Backup Unit  
SCSI Connector  
Non-Buffered 3.3V DIMM Socket  
J15  
RUBI Slot  
Interrupt Steering  
J15 J17  
J16  
J16  
RUBI Slot  
Interrupt Steering  
J14  
J17  
RUBI Slot  
Interrupt Steering  
Installation Steps  
This section provides an overview of installing the RAID controller. See the  
following sections for more information about each step.  
1
Unpack the PERC 3/SC controller and inspect for damage.  
Make sure all items are in the package. If damaged, call your Dell™  
original equipment manufacturer (OEM) support representative.  
2
3
4
5
Turn the computer off and remove the cover.  
Check the jumper settings on the PERC 3/SC controller.  
Set SCSI termination.  
Install the PERC 3/SC card.  
PERC 3/SC Hardware Installation  
103  
6
Connect the SCSI cables to SCSI devices.  
Make sure all cables are properly attached, and that the PERC 3/SC  
card is properly installed.  
7
8
Set the target IDs for the SCSI devices.  
Replace the computer cover and turn the power on.  
Be sure the SCSI devices are powered up before or at the same time as  
the host computer.  
9
Run the PERC 3 BIOS Configuration Utility or WebBIOS  
Configuration Utility to configure arrays and logical drives.  
10 Install software drivers for the desired operating systems.  
Step 1—Unpack  
Unpack and install the hardware in a static-free environment. The PERC 3  
is packed inside an anti-static bag between two sponge sheets. Remove the  
controller card and inspect it for damage. If the card appears damaged, or if  
any item listed below is missing, contact your Dell support representative.  
The PERC 3 controller also comes with:  
•
•
•
The PERC 3 User’s Guide  
A license agreement  
The PERC 3 configuration utilities diskette  
Step 2—Power Down  
Turn off the computer and remove the cover. Make sure the computer is  
disconnected from any networks before you install the controller card.  
104  
PERC 3/SC Hardware Installation  
Step 3—Set Jumpers  
Make sure the jumper settings on the PERC 3/SC card are correct. The  
jumpers and connectors are shown in Table 8-1.  
Table 8-1. PERC 3/SC Jumpers  
Connector  
Description  
Type  
J1  
J2  
J3  
SCSI bus termination enable control  
CPLD programming  
3-pin header  
10-pin header  
2-pin header  
Non-volatile random access memory  
(NVRAM) clear  
J4  
Serial erasable programmable read-  
only memory (EPROM)  
2-pin header  
J5  
J6  
J7  
J8  
Serial port  
3-pin header  
2-pin header  
2-pin header  
4-pin connector  
Write pending  
BIOS enable  
User activity light-emitting diode  
(LED)  
J9  
I2C connector  
4-pin header  
J10  
J11  
J13  
J15  
J16  
J17  
SCSI bus termination power  
Internal straddle-mount connector  
External SCSI connector  
RUBI slot interrupt steering  
RUBI slot interrupt steering  
RUBI slot interrupt steering  
2-pin header  
68-pin connector  
68-pin connector  
3-pin header  
3-pin header  
3-pin header  
J1 Termination Enable  
J1 is a three-pin header that specifies hardware or software control of SCSI  
termination. The default is OPEN. The settings are shown in Table 8-2.  
Table 8-2. J1 Termination Enable Settings  
Type of SCSI Termination  
J10 Setting  
Software control of SCSI termination using drive  
detection.  
Short pins 1-2  
PERC 3/SC Hardware Installation  
105  
Table 8-2. J1 Termination Enable Settings (continued)  
Type of SCSI Termination  
J10 Setting  
Short pins 2-3  
OPEN (default)  
Permanently disable all onboard SCSI termination.  
Permanently enable all onboard SCSI termination.  
J9 I2C Interface Connector  
J9 is a four-pin header that allows the i960RM core processor to serve as a  
master and slave device that resided on the I2C bus when used with the I2C  
bus interface unit. Attach a four-wire cable from J9 to the I2C Bus Interface  
Unit. Table 8-3 shows the J9 connector pinout.  
Table 8-3. J9 I2C Inter face Connector  
Pinout  
Pin  
1
Description  
SDA  
2
GND  
3
SCL  
4
VCC  
J5 Serial Port  
J5 is a 3-pin connector that attaches to a serial cable. Table 8-4 displays the  
pinout for J5.  
Table 8-4. J5 Serial Port Pinout  
Pin  
1
Description  
RXD  
2
TXD  
3
GND  
106  
PERC 3/SC Hardware Installation  
J8 Hard Disk LED  
J8 is a four-pin connector that attaches to a cable that connects to the hard  
disk light emitting diode (LED) mounted on the computer enclosure. The  
LED indicates data transfers.Table 8-5 shows the pinout for J8.  
Table 8-5. J8 Hard Disk LED  
Pin  
1
Description  
VCC through pull-up  
SCSI activity signal  
SCSI activity signal  
VCC through pull-up  
2
3
4
J10 Term Power  
J10 is a 2-pin jumper. The factory setting is pins 1-2 shorted. Pins 1-2 should  
always be shorted for J10 to enable onboard term power.  
J15 RUBI Slot Interrupt Steering  
J15 is a 3-pin jumper. You can short the pins for a standard PCI slot or a PCI  
RUBI slot. The default is to have the jumper installed to short pins 1 and 2.  
Table 8-6 shows the pinout for J15.  
Table 8-6. J15 RUBI Slot Interrupt Steering  
Short… For…  
Pins 1-2  
Pins 2-3  
Standard PCI slot (default)  
PCI RUBI slot  
PERC 3/SC Hardware Installation  
107  
J16, J17 RUBI Slot Interrupt Steering  
J16 and J17 are 3-pin jumpers. You can short them for a one-channel or two-  
channel motherboard. The default is OPEN; the pins are not shorted. Table  
8-7 shows the settings for shorting J16 and J17.  
Table 8-7. J16, J17 RUBI Slot Interrupt Steering  
Short…  
For…  
Pins 1-2 on both jumpers 2-channel motherboard RAID  
Pins 2-3 on both jumpers 1-channel motherboard  
Step 4—Set SCSI Termination  
The SCSI bus is an electrical transmission line and must be terminated  
properly to minimize reflections and losses. Termination should be set at  
each end of the SCSI cable(s), as shown in the following illustration:  
Figure 8-2. Example of SCSI Termination  
SCSI  
Terminator  
Termination on Controller  
Enabled  
SCSI Devices  
(Termination Disabled on Both)  
For a disk array, set SCSI bus termination so that removing or adding a  
SCSI device does not disturb termination. An easy way to do this is to  
connect the PERC 3/SC card to one end of the SCSI cable and to connect  
an external terminator module at the other end of the cable. The  
connectors between the two ends can connect SCSI devices. Disable  
termination on the SCSI devices. See the manual for each SCSI device to  
disable termination.  
108  
PERC 3/SC Hardware Installation  
SCSI Termination  
The SCSI bus is an electrical transmission line and it must be terminated  
properly to minimize reflections and losses. You complete the SCSI bus by  
setting termination at both ends.  
You can let PERC 3/SC automatically provide SCSI termination at one end  
of the SCSI bus. You can terminate the other end of the SCSI bus by  
attaching an external SCSI terminator module to the end of the cable or by  
attaching a SCSI device that internally terminates the SCSI bus at the end  
of the SCSI channel.  
Selecting a Terminator  
Use standard external SCSI terminators on a SCSI channel operating at 10  
MB/s or higher synchronous data transfer.  
Terminating Internal SCSI Disk Arrays  
Set the termination so that SCSI termination and termination power are  
intact when any physical drive is removed from a SCSI channel, as shown in  
Figure 8-3.  
PERC 3/SC Hardware Installation  
109  
Figure 8-3. Termination of Internal SCSI Disk Arrays  
Termination Enabled  
ID2  
ID1 – No Termination  
ID0 – Boot Drive  
No Termination  
PERC 3/SC  
SCSI ID 7  
110  
PERC 3/SC Hardware Installation  
Step 5—Install PERC 3/SC  
Choose a 3.3 V or 5 V PCI slot and align the PERC 3/SC card bus connector  
to the slot. Press down gently but firmly to make sure that the card is  
properly seated in the slot. The bottom edge of the controller card should  
be flush with the slot. Figure 8-4 shows PCI slots on a motherboard.  
Figure 8-4. PCI Slots on Motherboard  
PERC 3/SC Hardware Installation  
111  
Insert the PERC 3/SC card in a PCI slot, as shown in Figure 8-5. Screw the  
bracket to the computer chassis.  
Figure 8-5. Installation of PERC 3/SC Card into  
Motherboard  
Bracket Screw  
32-bit Slots  
(3.3 V)  
Edge of  
Mother Board  
64-bit Slots  
(5 V)  
112  
PERC 3/SC Hardware Installation  
Step 6—Connect SCSI Cables  
Connect the SCSI cables to SCSI devices, as shown in Figure 8-6. PERC  
3/SC provides two SCSI connectors:  
•
J11, the SCSI channel internal high-density 68-pin connector for Wide  
(16-bit) SCSI  
•
J13, the SCSI channel external ultra high-density 68-pin connector for  
Wide (16-bit) SCSI  
Make sure all cables are properly attached, and that the PERC 3/SC card is  
properly installed.  
Figure 8-6. Connecting SCSI Cables  
Pin 1  
J11  
J13  
Pin 1  
Connect SCSI Devices  
The cable length should not exceed 3 meters for Fast SCSI (10 MB/s)  
devices or single ended 1.5 meters for Ultra SCSI devices. The cable length  
can be up to 12 meters for low voltage differential (LVD) devices.  
The cable length should not exceed six meters for non-Fast SCSI devices.  
PERC 3/SC Hardware Installation  
113  
Perform the following steps to connect SCSI devices:  
1
Disable termination on any SCSI device that does not sit at the end of  
the SCSI bus.  
2
3
Configure all SCSI devices to supply termination power (TermPWR.)  
Set proper target IDs (TIDs) for all SCSI devices.  
Cable Suggestions  
System throughput problems can occur if SCSI cable use is not maximized.  
You sho uld :  
•
•
Use cables up to 12 meters for LVD devices.  
For single-ended SCSI devices, use the shortest SCSI cables (no more  
than 3 meters for Fast SCSI, no more than 1.5 meters for an 8-drive  
Ultra SCSI system, and no more than 3 meters for a 6-drive Ultra SCSI  
system.)  
•
•
•
Use active termination.  
Avoid clustering the cable nodes.  
Note that cable stub length should be no more than 0.1 meter (4  
inches.)  
•
•
•
Route SCSI cables carefully.  
Use high impedance cables.  
Do not mix cable types. (Choose either flat or rounded and shielded or  
non-shielded.)  
•
Note that ribbon cables have fairly good cross-talk rejection  
characteristics.  
Step 7—Set Target IDs  
Set target identifiers (TIDs) on the SCSI devices. Each device in a specific  
SCSI channel must have a unique TID in that channel. Non-disk devices  
(CD or tapes) should have unique SCSI IDs regardless of the channel where  
they are connected. See the documentation for each SCSI device to set the  
TIDs.  
114  
PERC 3/SC Hardware Installation  
The PERC 3/SC controller automatically occupies TID 7 in the SCSI  
channel. Eight-bit SCSI devices can use the TIDs from 0 to 6 only. 16-bit  
devices can use the TIDs from 0 to 15. The arbitration priority for a SCSI  
device depends on its TID. Table 8-8 shows the target IDs.  
Table 8-8. Target IDs  
Priority  
TID  
Highest  
Lowest  
15 14 ...  
7
6
5
...  
2
1
0
9
8
NOTICE: Non-disk devices (CD-ROM or tapes) should have unique SCSI IDs  
regardless of the channel they are connected to. ID 0 cannot be used for non-  
disk devices because they are limited to IDs 1 through 6. There is a limit of six  
IDs for non-disk devices per controller.  
Step 8—Power On Host System  
Replace the computer cover and reconnect the alternating current (AC)  
power cords. Turn power on to the host computer. Set up the power supplies  
so that the SCSI devices are powered up at the same time as or before the  
host computer. If the computer is powered up before a SCSI device, the  
device might not be recognized.  
During boot, the PERC 3 BIOS message appears:  
PowerEdge Expandable RAID Controller BIOS Version x.xx  
date  
Copyright (c) Dell Computer Corporation  
Firmware Initializing... [Scanning SCSI Device...(etc.)...]  
The firmware takes several seconds to initialize. During this time the  
adapter scans the SCSI channel. When ready, the following appears:  
HA –0 (Bus 1 Dev 6) Type: PERC 3/SC Standard FW x.xx SDRAM=  
xxxMB  
0 Logical Drives found on the Host Adapter  
0 Logical Drive(s) handled by BIOS  
Press <Ctrl><M> to run PERC 3 BIOS Configuration Utility  
Or press <Ctrl><H> to run WebBIOS  
PERC 3/SC Hardware Installation  
115  
The PERC 3 BIOS Configuration Utility prompt times out after several  
seconds. The PERC 3 host adapter (controller) number, firmware version,  
and cache synchronized dynamic random access memory (SDRAM) size  
display in the second portion of the BIOS message. The numbering of the  
controllers follows the PCI slot scanning order used by the host  
motherboard.  
Step 9—Run PERC 3 BIOS Configuration Utility or WebBIOS  
Utility  
Press <Ctrl><M> to run the PERC 3 BIOS Configuration Utility or  
<Ctrl><H> to run the WebBIOS Configuration Utility. See Chapter 11  
"PERC 3 BIOS Configuration Utility" for additional information about  
running the BIOS Configuration Utility. WebBIOS is an HTML-based  
utility. See the WebBIOS Configuration Utility Guide for additional  
information about running the WebBIOS Configuration Utility.  
Step 10—Install Operating System Software Drivers  
See the PERC 3 Operating System Driver Installation Guide for additional  
information about installing the drivers for Windows NT®, Windows®  
2000, Novell® NetWare® 5.x, 6.x, and Red Hat Linux 7.x.  
116  
PERC 3/SC Hardware Installation  
SE CT ION 9  
PERC 3/DC or PERC  
3/DCL Hardware  
Installation  
Requirements  
PERC 3/DC Card Layout  
PERC 3/DCL Card Layout  
Installation Steps  
Replacing a PERC 3/DC Containing a BC Chip with a  
PERC 3/DC Containing a BE Chip  
Requirements  
This chapter describes the installation procedures for the PERC 3/DC, and  
PERC 3/DCL boards. It also contains the procedure for replacing the PERC  
3/DC that has a BC chip with a PERC 3/DC that has a BE chip. You must  
have the following:  
•
•
A PERC 3/DC or PERC 3/DCL controller  
NOTE: The PERC 3/DC  
or PERC 3/DCL  
controller must be  
installed in a PCI  
expansion slot.  
A host computer with an available 32- or 64-bit, 5 V peripheral  
component interconnect (PCI) expansion slot  
•
•
The PERC 3/DC or PERC 3/DCL installation diskettes  
The necessary SCSI cables (This depends on the number and type of  
SCSI devices to be attached.)  
•
Low-voltage differential (LVD) or 160M SCSI hard drives  
Dell™ strongly recommends that you have an uninterruptible power supply  
(UPS) for the entire system.  
NOTE: The PERC 3/DC  
has a battery option; the  
PERC 3/DCL does not.  
Optional Equipment  
You may also want to install SCSI cables that connect PERC 3/DC or PERC  
3/DCL to external SCSI devices.  
118  
PERC 3/DC or PERC 3/DCL Hardware Installation  
PERC 3/DC Card Layout  
Figure 9-1. PERC 3/DC Card L ayout  
J4 Channel 0 Internal  
High-Density 68-pin  
SCSI Connector  
J5 Channel 1 Internal  
High-Density 68-pin  
SCSI Connector  
J2  
J3  
J7 Channel 0  
External  
Ultra-High  
Density  
68-pin SCSI  
Connector  
J11  
J10  
J9  
J8  
J15 Channel 1  
External  
Ultra-High  
Density  
68-pin SCSI  
Connector  
J13  
J16  
J14  
J17  
J18  
J19  
J16 Channel 0 Terminator Power Enable  
J18 Channel 1 Power Enable  
J19  
J8 Serial EPROM  
J9 Onboard BIOS Enable J14 SCSI Activity  
J10 NVRAM Clear  
J13 Dirty Cache  
J17 I2C  
J11 Serial Port Connector  
PERC 3/DC or PERC 3/DCL Hardware Installation  
119  
PERC 3/DCL Card Layout  
Figure 9-2. PERC 3/DCL Card L ayout  
J4 Channel 0 Internal  
High-Density 68-pin  
SCSI Connector  
J5 Channel 1 Internal  
High-Density 68-pin  
SCSI Connector  
J2  
J3  
J7 Channel 0  
External  
Ultra-High  
Density  
68-pin SCSI  
Connector  
J11  
J10  
J9  
J8  
J15 Channel 1  
External  
Ultra-High  
Density  
J13  
J16  
J18  
J14  
J17  
68-pin SCSI  
Connector  
J19  
J16 Channel 0 Terminator Power Enable  
J18 Channel 1 Power Enable  
J19  
J8 Serial EPROM  
J9 Onboard BIOS Enable J14 SCSI Activity  
J10 NVRAM Clear  
J13 Dirty Cache  
J17 I2C  
J11 Serial Port Connector  
Installation Steps  
The steps for installing the RAID controller are shown below. See the  
following pages for more information about each step.  
1
2
3
4
5
6
Unpack the PERC 3/DC or PERC 3/DCL.  
Power off the host system.  
Set jumpers.  
Set SCSI termination.  
Install the PERC 3/DC or PERC 3/DCL controller.  
Select and set target IDs for SCSI devices.  
120  
PERC 3/DC or PERC 3/DCL Hardware Installation  
7
8
9
Connect SCSI devices.  
Power on the host system.  
Run the BIOS Configuration Utility or the WebBIOS Configuration  
Utility.  
10 Install the operating system software driver.  
Step 1—Unpack the PERC 3/DC or PERC 3/DCL  
Unpack and install the hardware in a static-free environment. Remove the  
controller card and inspect it for damage. If the card appears damaged,  
please contact your Dell support representative.  
Step 2—Power Down  
Turn off the computer and remove the AC power cord. Ensure that the  
computer is disconnected from any networks before installing the controller.  
Remove the system cover. For information regarding removal of system  
covers, please consult the system documentation.  
Step 3—Set Jumpers  
Make sure the jumper settings on the PERC 3/DC or PERC 3/DCL card are  
correct. The jumpers and connectors are shown in Table 9-1.  
Table 9-1. Jumper Settings  
Connector  
Description  
Type  
J2  
J3  
J4  
J5  
J7  
J8  
Channel 1 termination enable  
Channel 0 termination enable  
Channel 0 internal Wide SCSI  
Channel 1 internal Wide SCSI  
Channel 0 external Wide SCSI  
3-pin header  
3-pin header  
68-pin connector  
68-pin connector  
68-pin connector  
2-pin header  
Serial erasable programmable read-  
only memory (EEPROM) port  
J9  
Onboard BIOS enable  
2-pin header  
J10  
Non-volatile random access memory  
(NVRAM) clear  
2-pin connector  
J11  
Serial port connector  
3-pin header  
PERC 3/DC or PERC 3/DCL Hardware Installation  
121  
Table 9-1. Jumper Settings (continued)  
Connector  
Description  
Type  
J13  
J14  
J15  
J16  
J17  
J18  
Dirty cache (Write Pending) LED  
SCSI activity LED  
2-pin header  
4-pin header  
68-pin connector  
2-pin header  
4-pin header  
2-pin header  
Channel 1 external Wide SCSI  
Channel 0 TERMPWR enable  
I2C connector  
Channel 1 TERMPWR enable  
J2, and J3 Termination Enable  
J2, and J3 are 3-pin connectors that set the SCSI termination for each SCSI  
channel. The default is OPEN. Table 9-2 displays the termination enable  
settings.  
Table 9-2. J2 and J3 Termination Enable Settings  
Jumper  
SCSI  
SCSI  
SCSI  
SCSI  
Channel  
Termination  
Controlled by  
Software  
Termination  
Always  
Disabled  
Termination  
Always  
Enabled  
J2  
J3  
0
1
Short pins 1-2  
Short pins 2-3  
OPEN  
(default)  
Short pins 1-2  
Short pins 2-3  
OPEN  
(default)  
J9 Onboard BIOS Enable  
J9 is a 2-pin connector that enables or disables the PERC 3/DC or PERC  
3/DCL onboard BIOS. The onboard BIOS should be enabled (J9  
unjumpered) for normal board operation. Unjumpered is the default. Table  
9-3 displays the settings for J9.  
Table 9-3. J9 Onboard BIOS Enable Settings  
J9 Setting  
Unjumpered  
Jumpered  
Onboard BIOS Status  
Enabled (default)  
Disabled  
122  
PERC 3/DC or PERC 3/DCL Hardware Installation  
J10 NVRAM Clear  
J10 is a 2-pin connector used to clear the memory from the NVRAM, which  
stores RAID configuration information. Table 9-4 displays the pinout for  
J10.  
Table 9-4. J10 NVRAM Clear Pinout  
Pin  
1
Description  
Signal  
2
GND  
J11 Serial Port  
J11 is a 3-pin header that attaches to a serial cable. Table 9-5 and Figure 9-3  
display the pinout.  
Table 9-5. J11 Serial Port Pinout  
Pin  
1
Description  
Receive data  
Transmit data  
Ground  
2
3
Figure 9-3. J11 Serial Port Pinout  
J11  
3 Ground  
2 Transmit Data  
1 Receive Data  
J13 Dirty Cache LED  
J13 is a two-pin connector for an LED mounted on the computer enclosure.  
The LED indicates when the data in the cache has yet to be written to the  
storage devices. Table 9-6 displays the J13 pinout.  
PERC 3/DC or PERC 3/DCL Hardware Installation  
123  
Table 9-6. J13 Dirty Cache LED Pinout  
Pin  
1
Description  
Signal pulled high  
Dirty cache signal  
2
J14 SCSI Activity LED  
J14 is a four-pin connector for an LED mounted on the computer enclosure.  
Table 9-7 displays the J14 pinout.  
Table 9-7. J14 SCSI Activity Pinout  
Pin  
1
Description  
VCC  
2
SCSI activity signal  
SCSI activity signal  
VCC  
3
4
J16 and J18 TERMPWR Enable  
J16, and J18 are 2-pin connectors that enable termination power to the  
SCSI bus for each SCSI channel. The default is pins 1 and 2 jumpered.  
Table 9-8 displays the settings for J16 and J18.  
Table 9-8. J16 and J18 TERMPWR Enable Settings  
Jumper  
Termination  
Power  
Settings  
Channel  
J16  
0
Short pins 1-2 to have the PCI bus on the host  
computer provide TermPWR. This is the default  
setting. Leave open to let the SCSI bus provide  
TermPWR.  
J18  
1
Short pins 1-2 to have the PCI bus on the host  
computer provide TermPWR. This is the default  
setting. Leave open to let the SCSI bus provide  
TermPWR.  
124  
PERC 3/DC or PERC 3/DCL Hardware Installation  
J17 I2C Connector  
J17 is a 4-pin header. Table 9-9 displays the J17 pinout.  
Table 9-9. J17 I2C Connector Pinout  
Pin  
1
Description  
Data  
2
GND  
3
Clock  
4
Power (fused)  
Step 4—Set SCSI Termination  
Each PERC 3/DC or PERC 3/DCL SCSI channel can be individually  
configured for termination enable mode by setting the J2 and J3 jumpers.  
You must terminate the SCSI bus properly. Set termination at both ends of  
the SCSI cable. The SCSI bus is an electrical transmission line and must be  
terminated properly to minimize reflections and losses. Termination should  
be set at each end of the SCSI bus, as shown in the figures.  
Terminating Internal SCSI Disk Arrays  
Set the termination so that SCSI termination and termination power are  
intact when any disk drive is removed from a SCSI channel, as shown in  
Figure 9-4.  
PERC 3/DC or PERC 3/DCL Hardware Installation  
125  
Figure 9-4. Termination of Internal SCSI Disk Arrays for  
PERC 3/DC and 3/DCL  
Internal SCSI Drives  
ID0  
Boot Drive  
No Term.  
ID1  
No Term.  
ID2  
Term.  
Enabled  
For a disk array, set SCSI bus termination so that removing or adding a  
SCSI device does not disturb termination. The connectors between the two  
ends of the channel can connect SCSI devices that have termination  
enabled. See the manual for each SCSI device to disable termination.  
The PERC 3/DC or PERC 3/DCL can provide SCSI termination  
automatically at one end of the SCSI bus. The other end of the SCSI bus  
can be terminated by attaching an external SCSI bus terminator module to  
the end of the cable, or by attaching a SCSI device that internally  
terminates the SCSI bus at the end of the SCSI channel.  
Step 5—Install the PERC 3/DC or PERC 3/DCL Controller  
Choose a 3.3 V or 5 V PCI slot and align the PERC 3/DC or PERC 3/DCL  
bus connector to the slot. Press down gently but firmly to make sure that the  
card is properly seated in the slot. The bottom edge of the controller card  
should be flush with the slot.  
Insert the PERC 3/DC or PERC 3/DCL card into a 64-bit PCI slot, as  
shown in Figure 9-5. Screw the bracket to the computer frame.  
126  
PERC 3/DC or PERC 3/DCL Hardware Installation  
Figure 9-5. Installation of the PERC 3/DC and 3/DCL  
Bracket Screw  
32-bit Slots  
(5 V)  
Edge of  
Mother Board  
64-bit Slots  
(5 V)  
PERC 3/DC or PERC 3/DCL Hardware Installation  
127  
Step 6—Select and Set Target IDs for SCSI Devices  
Set target identifiers (TIDs) on the SCSI devices. Each device in a specific  
SCSI channel must have a unique TID in that channel. See the  
documentation for each SCSI device to set the TIDs. The PERC 3/DC or  
PERC 3/DCL controller automatically occupies TID 7 in the SCSI channel.  
8-bit SCSI devices can use the TIDs from 0 to 6 only. 16-bit devices can use  
the TIDs from 0 to 15. The arbitration priority for a SCSI device depends  
on its TID.  
Table 9-10. Target IDs for SCSI Devices  
Priority  
Highest  
Lowest  
15 14 ...  
TID  
7
6
5
...  
2
1
0
9
8
Step 7—Connect SCSI Cables  
Connect SCSI cables to SCSI devices. PERC 3/DC or PERC 3/DCL  
provides internal SCSI connectors, J4 and J5, which are the SCSI channel  
internal high-density 68-pin connectors for Wide (16-bit) SCSI. PERC  
3/DC or PERC 3/DCL also provides two external SCSI connectors: J7 and  
J15, the SCSI channel external ultra high-density 68-pin connectors for  
Wide (16-bit) SCSI. Make sure pin 1 on the card is connected to pin 1 of  
the cable.  
128  
PERC 3/DC or PERC 3/DCL Hardware Installation  
Figure 9-6. Connecting SCSI Cables to PERC 3/DC and  
3/DCL  
Pin 1  
Pin 1  
Connect SCSI Devices  
When connecting SCSI devices:  
1
Disable termination on any SCSI device that is not at the end of the  
SCSI bus.  
If the cable is terminated, do not terminate SCSI devices.  
Configure all SCSI devices to supply TermPWR.  
Set proper target IDs (TIDs) for all SCSI devices.  
2
3
4
The cable length should not exceed 12 meters for LVD and 160M  
SCSI devices.  
Cable Suggestions  
SCSI disk subsystem throughput can impaired if SCSI cabling is not  
optimized. Dell suggests using the following:  
•
•
Actively terminated cables  
Dell-approved cables  
PERC 3/DC or PERC 3/DCL Hardware Installation  
129  
Step 8—Power On Host System  
Replace the computer cover and reconnect the alternating current (AC)  
power cords. Power on the host computer. Ensure that all SCSI devices are  
connected, including SCSI and power cables. If the computer is powered up  
before a SCSI device, the device might not be recognized.  
During boot, the PERC 3/DC or PERC 3/DCL BIOS message appears: (For  
simplicity, only the PERC 3/DC message is shown.)  
PowerEdge Expandable RAID Controller BIOS Version xwxx  
date  
Copyright (c) Dell Computer Corporation  
Firmware Initializing... [Scanning SCSI Device...(etc.)...]  
The firmware takes several seconds to initialize. During this time the  
adapter will scan the SCSI channel. When ready, the following appears:  
HA –0 (Bus 1 Dev 6) Type: PERC 3/DC Standard FW xwxx SDRAM=  
128MB  
Battery Module is Present on Adapter  
0 Logical Drives found on the Host Adapter  
0 Logical Drive(s) handled by BIOS  
Press <Ctrl><H> to run WebBIOS Configuration Utility  
The PERC 3 BIOS Configuration Utility prompt times out after several  
seconds. The PERC 3/DC or PERC 3/DCL host adapter (controller)  
number, firmware version, and cache SDRAM size display in the second  
portion of the BIOS message. The numbering of the controllers follows the  
PCI slot scanning order used by the host motherboard.  
Step 9—Run the PERC 3 BIOS Configuration Utility or  
WebBIOS Configuration Utility  
Press <Ctrl><M> to run the PERC 3 BIOS Configuration Utility or  
<Ctrl> <H> to run the PERC 3 WebBIOS Configuration Utility.  
WebBIOS is an HTML-based configuration utility. See Chapter 11 "PERC  
3 BIOS Configuration Utility" for information about running the BIOS  
Configuration Utility. See the WebBIOS Configuration Utility Guide for  
more information about running the WebBIOS Configuration Utility.  
130  
PERC 3/DC or PERC 3/DCL Hardware Installation  
Step 10—Install Operating System Software Driver  
See the documentation for the PERC 3/DC or PERC 3/DCL operating  
systems driver for additional information about installing the drivers for  
Windows NT, Windows 2000, Novell NetWare 5.x, 6.x, and Red Hat Linux  
7.x.  
Replacing a PERC 3/DC Containing a  
BC Chip with a PERC 3/DC Containing  
a BE Chip  
Using Driver 5.22.1 or 5.22.2 and Firmware 161J or 161N  
Perform the following steps to replace the PERC 3/DC card if you are using  
the 5.22.1 or 5.22.2 driver, and 161J or 161N firmware.  
1
If the driver is not updated to 5.30 or later before you switch  
controllers, shut down the machine that will have its controller  
replaced.  
2
3
4
Replace the controller (using the procedures in this chapter.)  
Boot to Windows.  
When logging into Windows, cancel the New Hardware Device  
Found prompts.  
5
6
7
8
Power down the machine.  
Power on the machine again.  
Log into Windows.  
Proceed to the Windows Device Manager.  
There will be a yellow marker on the RAID controller.  
9
Choose the update driver by selecting Display a list of the Known  
drivers for this device and specifying the location of the driver.  
Using Driver 5.30 and Firmware 1.70  
Perform the following steps if the driver is updated to 5.30 or later before  
you replace the controller, and the firmware on the new controller is  
updated to 1.70.  
PERC 3/DC or PERC 3/DCL Hardware Installation  
131  
1
2
Upgrade the existing driver from 5.22.1 or 5.22.2 to 5.30 or later.  
Shut down the machine in which you are going to replace the  
controller.  
3
4
Replace the controller (using the procedures in this chapter.)  
Boot to Windows.  
When you log into Windows, New Device Prompts appears.  
Load the driver if prompted.  
5
132  
PERC 3/DC or PERC 3/DCL Hardware Installation  
S EC T IO N 1 0  
PERC 3/QC Hardware  
Installation  
Requirements  
PERC 3/QC Card Layout  
Installation Steps  
Requirements  
This chapter describes the installation procedures for the PERC 3/QC  
board. You must have the following:  
•
•
A PERC 3/QC controller  
A host computer with an available 5 V, 32- or 64-bit peripheral  
component interconnect (PCI) expansion slot  
•
•
The PERC 3/QC installation diskettes  
The necessary SCSI cables (This depends on the number and type of  
SCSI devices to be attached.)  
•
Fast, Ultra, Ultra II, and 160M SCSI hard drives  
Dell™ strongly recommends that you have an uninterruptible power supply  
(UPS) for the entire system.  
Optional Equipment  
You may also want to install SCSI cables that connect PERC 3/QC to  
external SCSI devices.  
134  
PERC 3/QC Hardware Installation  
PERC 3/QC Card Layout  
Figure 10-1 displays the PERC 3/QC card and jumpers.  
Figure 10-1. PERC 3/QC Card L ayout  
PERC 3/QC Hardware Installation  
135  
Installation Steps  
This section provides an overview of installing the RAID controller. See the  
following sections for more information about each step.  
1
Unpack the PERC 3/QC controller and inspect for damage.  
Make sure all items are in the package. If damaged, call your Dell  
original equipment manufacturer (OEM) support representative.  
2
3
4
5
6
7
8
Turn the computer off and remove the cover.  
Check the jumper settings on the PERC 3/QC controller.  
Set SCSI termination.  
Install the PERC 3/QC card.  
Connect the SCSI cables to SCSI devices.  
Set the target IDs for the SCSI devices.  
Replace the computer cover and turn the power on.  
Be sure the SCSI devices are powered up before or at the same time as  
the host computer.  
9
Run the PERC 3/QC Basic Input/Output System (BIOS)  
Configuration Utility or WebBIOS Configuration Utility.  
10 Install software drivers for the desired operating systems.  
Step 1—Unpack the PERC 3/QC Controller  
Unpack and install the hardware in a static-free environment. The PERC  
3/QC is packed inside an anti-static bag between two sponge sheets.  
Remove the controller card and inspect it for damage. If the card appears  
damaged, or if any item listed below is missing, contact your Dell support  
representative. The PERC 3/QC card comes with:  
•
•
•
The PERC 3 User’s Guide  
A license agreement  
The PERC 3/QC Configuration Utilities diskette  
136  
PERC 3/QC Hardware Installation  
Step 2—Power Down  
Turn off the computer and remove the cover. Make sure the computer is  
turned off and disconnected from any networks before installing the  
controller.  
Step 3—Set Jumpers  
Make sure the jumper settings on the PERC 3/QC card are correct. Table  
10-1 displays the jumpers and connectors.  
Table 10-1. PERC 3/QC Jumper Settings  
Connector  
Description  
Type  
J1  
J2  
J3  
J4  
J5  
J6  
Channel 1 internal Wide SCSI  
Channel 0 termination enable  
Channel 1 termination enable  
Channel 0 internal Wide SCSI  
Channel 2 termination enable  
68-pin connector  
3-pin header  
3-pin header  
68-pin connector  
3-pin header  
4-pin header  
SCSI activity light emitting diode  
(LED)  
J7  
Channel 3 termination enable  
Channel 0 TERMPWR enable  
Channel 1 TERMPWR enable  
Channel 2 TERMPWR enable  
Channel 3 TERMPWR enable  
Channel 0/1 external Wide SCSI  
Serial port connector  
3-pin header  
J9  
2-pin header  
J10  
J11  
J12  
J13  
J14  
J17  
J18  
J19  
J22  
J23  
2-pin header  
2-pin header  
2-pin header  
Dual 68-pin connector  
9-pin connector  
2-pin connector  
2-pin header  
Dirty cache LED  
Serial EEPROM port  
Onboard BIOS enable  
2-pin header  
Channel 2/3 external Wide SCSI  
External battery connector  
Dual 68-pin connector  
5-pin connector  
PERC 3/QC Hardware Installation  
137  
J2, J3, J5, and J7 Termination Enable  
J2, J3, J5, and J7 are 3-pin connectors that set the SCSI termination for each  
SCSI channel. The Dell default is termination always enabled (OPEN.)  
Table 10-2 displays the pinout.  
Table 10-2. J2, J3, J5 and J7 Termination Enable Pinout  
Jumper SCSI  
SCSI  
SCSI  
Termination  
Always  
SCSI  
Termination  
Always Enabled  
Channel Termination  
Controlled by  
Software  
Disabled  
J2  
J3  
J5  
J7  
0
1
2
3
Short pins 1-2  
Short pins 1-2  
Short pins 1-2  
Short pins 1-2  
Short pins 2-3  
Short pins 2-3  
Short pins 2-3  
Short pins 2-3  
OPEN (default)  
OPEN (default)  
OPEN (default)  
OPEN (default)  
J9, J10, J11, and J12 TERMPWR Enable  
J9, J10, J11, and J12 are 2-pin connectors that enable TERMPWR to the  
SCSI bus for each SCSI channel. The default is pins 1 and 2 shorted. Table  
10-3 displays the pinout.  
Table 10-3. J9, J10, J11 and J12 TERMPWR Enable Pinout  
Jumper  
Term.  
Settings  
Power  
Channel  
J9  
0
1
2
3
Short pins 1-2 to have the host PCI bus provide  
TermPWR. This is the factory setting. Leave open to let  
the SCSI bus provide TermPWR.  
J10  
Short pins 1-2 to have the host PCI bus provide  
TermPWR. This is the factory setting. Leave open to let  
the SCSI bus provide TermPWR.  
J11  
Short pins 1-2 to have the host PCI bus provide  
TermPWR. This is the factory setting. Leave open to let  
the SCSI bus provide TermPWR.  
J12  
Short pins 1-2 to have the host PCI bus provide  
TermPWR. This is the factory setting. Leave open to let  
the SCSI bus provide TermPWR.  
138  
PERC 3/QC Hardware Installation  
J14 Serial Port  
J14 attaches to a serial cable. Figure 10-2 and Table 10-4 show the pinout for  
J14.  
Figure 10-2. J14 Serial Port Diagram  
2
4
6
8
1
3
5
7
9
Table 10-4. J14 Serial Port Pinout  
Pin  
1
Description  
Carrier detect  
Receive data  
Transmit data  
Data terminal ready  
Ground  
Pin  
2
Description  
Data set ready  
Request to send  
Clear to send  
Ring indicator  
CUT  
3
4
5
6
7
8
9
10  
J17 Dirty Cache LED  
J17 is a two-pin connector for an LED mounted on the computer enclosure.  
The LED indicates when the data in the cache has yet to be written to the  
storage devices. Table 10-5 displays the pinout for J17.  
Table 10-5. J17 Dirty Cache LED Pinout  
Pin  
1
Description  
Signal pulled high  
Dirty cache signal  
2
J19 Onboard BIOS Enable  
J19 is a 2-pin connector which enables or disables PERC 3/QC onboard  
BIOS. The onboard BIOS should be enabled (J19 unjumpered) for normal  
board position. Unjumpered is the default. Table 10-6 displays the J19  
settings.  
PERC 3/QC Hardware Installation  
139  
Table 10-6. J19 Onboard BIOS Enable Settings  
J19 Setting  
Unjumpered  
Jumpered  
Onboard BIOS Status  
Enabled  
Disabled  
J23 External Battery  
J23 is a 5-pin connector that attaches to the optional battery pack. Table 10-  
7 displays the J23 pinout.  
Table 10-7. J23 External Batter y Pinout  
Pin  
1
Description  
+BATT Terminal (red wire)  
Thermistor (white wire)  
-BATT Terminal (black wire)  
BATDQ (no wire)  
2
3
4
5
Ground (no wire)  
Step 4—Set SCSI Termination  
You must terminate the SCSI bus properly. Set termination at both ends of  
the SCSI cable. The SCSI bus is an electrical transmission line and must be  
terminated properly to minimize reflections and losses. Termination should  
be set at each end of the SCSI cable(s), as shown below. Termination is  
always enabled, regardless of the configuration. However, you can override  
this setting by setting another state. The Dell default is termination by  
jumper. Figure 10-3 displays an example of termination.  
140  
PERC 3/QC Hardware Installation  
Figure 10-3. Example of Termination  
SCSI  
Terminator  
Termination on Controller  
Enabled  
SCSI Devices  
(Termination Disabled on Both)  
For a disk array, set SCSI bus termination so that removing or adding a  
SCSI device does not disturb termination. An easy way to do this is to  
connect the PERC 3/QC card to one end of the SCSI cable and to connect  
an external terminator module at the other end of the cable. The  
connectors between the two ends can connect SCSI devices. Disable  
termination on the SCSI devices. See the manual for each SCSI device to  
disable termination.  
SCSI Termination  
The SCSI bus is an electrical transmission line and it must be terminated  
properly to minimize reflections and losses. You complete the SCSI bus by  
setting termination at both ends.  
You can let PERC 3/QC automatically provide SCSI termination at one end  
of the SCSI bus. You can terminate the other end of the SCSI bus by  
attaching an external SCSI terminator module to the end of the cable or by  
attaching a SCSI device that internally terminates the SCSI bus at the end  
of the SCSI channel.  
Selecting a Terminator  
Use standard external SCSI terminators on a SCSI channel operating at 10  
MB/s or higher synchronous data transfer.  
PERC 3/QC Hardware Installation  
141  
Terminating Internal SCSI Disk Arrays  
Set the termination so that SCSI termination and termination power are  
intact when any disk drive is removed from a SCSI channel, as shown in  
Figure 10-4.  
Figure 10-4. Termination of Internal SCSI Disk Arrays  
Internal SCSI Drives  
ID0  
Boot Drive  
No Termination  
ID1  
No Termination  
ID2  
Termination  
Enabled  
142  
PERC 3/QC Hardware Installation  
Step 5—Install PERC 3/QC  
Choose a 3.3 V or 5 V PCI slot and align the PERC 3/QC bus connector to  
the slot. Press down gently but firmly to make sure that the card is properly  
seated in the slot. The bottom edge of the controller card should be flush  
with the slot. Figure 10-5 displays the PCI slots on a motherboard.  
Figure 10-5. PCI Slots on Motherboard  
PERC 3/QC Hardware Installation  
143  
Insert the PERC 3/QC card into a PCI slot as shown in Figure 10-6. Screw  
the bracket to the computer frame.  
Figure 10-6. Installation of the PERC 3/QC Card  
Bracket Screw  
32-bit Slots  
(3.3 V)  
Edge of  
Mother Board  
64-bit Slots  
(5 V)  
144  
PERC 3/QC Hardware Installation  
Step 6—Connect SCSI Cables  
Connect SCSI cables to SCSI devices. PERC 3/QC provides two internal  
SCSI connectors, J1 (channel 1) and J4 (channel 0), which are the SCSI  
channel internal high-density 68-pin connectors for Wide (16-bit) SCSI.  
PERC 3/QC provides four external SCSI connectors: J13 (channels 0 and 1)  
and J22 (channels 2 and 3), the SCSI channel external ultra high-density  
68-pin connectors for Wide (16-bit) SCSI. Only one cable should be  
connected to any available channel.  
Make sure that pin 1 on the cable is connected to pin 1 on the PERC 3/QC  
card. Figure 10-7 displays SCSI cables being connected to the card.  
Figure 10-7. Connecting SCSI Cables  
Pin 1  
J1  
J4  
J13  
J22  
Connect SCSI Devices  
The cable length should not exceed three meters for Fast SCSI (10 MB/s)  
devices or three meters for a four drive Ultra2 SCSI drive system. The cable  
length can be up to 12 meters for low-voltage differential (LVD) devices.  
The cable length should not exceed six meters for non-Fast SCSI devices.  
Perform the following steps to connect SCSI devices:  
1
Disable termination on any SCSI device that does not sit at the end of  
the SCSI bus.  
2
3
Configure all SCSI devices to supply TermPWR.  
Set proper target IDs (TIDs) for all SCSI devices.  
PERC 3/QC Hardware Installation  
145  
Cable Suggestions  
System throughput problems can occur if SCSI cable use is not maximized.  
You sho uld :  
•
•
Use cables up to 12 meters for LVD devices.  
For single ended SCSI devices, use the shortest SCSI cables (no more  
than 3 meters for Fast SCSI, no more than 1.5 meters for an 8-drive  
Ultra2 SCSI system and no more than 3 meters for a 6-drive Ultra2  
SCSI system.)  
•
•
Use active termination.  
Use only Dell-approved cables.  
Step 7—Set Target IDs  
Set target identifiers (TIDs) on the SCSI devices. Each device in a specific  
SCSI channel must have a unique TID in that channel. See the  
documentation for each SCSI device to set the TIDs. The PERC 3/QC  
controller automatically occupies TID 7 in the SCSI channel. Eight-bit  
SCSI devices can only use the TIDs from 0 to 6. 16-bit devices can use the  
TIDs from 0 to 15. The arbitration priority for a SCSI device depends on its  
TID.  
Table 10-8. Target IDs  
Priority  
Highest  
Lowest  
15 14 ...  
TID  
7
6
5
...  
2
1
0
9
8
Step 8—Power on Host System  
Replace the computer cover and reconnect the AC power cords. Turn power  
on to the host computer. Set up the power supplies so that the SCSI devices  
are powered up at the same time as or before the host computer. If the  
computer is powered up before a SCSI device, the device might not be  
recognized.  
During boot, the PERC 3/QC BIOS message appears:  
PowerEdge Expandable RAID Controller BIOS Version x.xx  
date  
Copyright (c) Dell Computer Corporation  
146  
PERC 3/QC Hardware Installation  
Firmware Initializing... [Scanning SCSI Device...(etc.)...]  
The firmware takes several seconds to initialize. During this time the  
adapter will scan the SCSI channel. When ready, the following appears:  
HA –0 (Bus 1 Dev 6) Type: PERC 3/QC Standard FW x.xx SDRAM=  
xxxMB  
Battery Module is Present on Adapter  
0 Logical Drives found on the Host Adapter  
0 Logical Drive(s) handled by BIOS  
Press <Ctrl><M> to run PERC 3 BIOS Configuration Utility  
Or press <Ctrl><H> to run WebBIOS  
The BIOS Configuration Utility prompt times out after several seconds.  
The PERC 3/QC host adapter (controller) number, firmware version, and  
cache SDRAM size are displayed in the second portion of the BIOS  
message. The numbering of the controllers follows the PCI slot scanning  
order used by the host motherboard.  
Step 9—Run the PERC 3 BIOS Configuration Utility or  
WebBIOS Configuration Utility  
Press <Ctrl><M> to run the PERC 3 BIOS Configuration Utility or  
<Ctrl> <H> to run the WebBIOS Configuration Utility. See Chapter 11  
"PERC 3 BIOS Configuration Utility" for additional information about  
running the PERC 3 BIOS Configuration Utility. WebBIOS is an HTML-  
based configuration utility. See the WebBIOS Configuration Utility Guide  
for additional information about running WebBIOS.  
Step 10—Install Operating System Driver  
See the PERC 3 Operating System Driver Installation Guide for additional  
information about installing the drivers for Windows NT®, Windows®  
2000, Novell® NetWare®, and Red Hat Linux.  
PERC 3/QC Hardware Installation  
147  
148  
PERC 3/QC Hardware Installation  
S EC T IO N 1 1  
PERC 3 BIOS  
Configuration Utility  
Configuration On Disk  
Starting the PERC 3 BIOS Configuration Utility  
BIOS Configuration Utility Menu Options  
Configuring Arrays and Logical Drives  
Choosing the Configuration Method  
Designating Drives as Hot Spares  
Using Easy Configuration  
Using New Configuration  
Using View/Add Configuration  
Initializing Logical Drives  
Deleting Logical Drives (Random Array Deletion)  
Formatting Physical Drives  
Rebuilding Failed Disk Drives  
Using a Pre-loaded SCSI Drive “As-is”  
Exiting the PERC 3 BIOS Configuration Utility  
Setting Hardware Termination  
Clustering  
The PERC 3 BIOS Configuration Utility configures disk arrays and logical  
drives. Because the utility resides in the PERC 3 BIOS, its operation is  
independent of the operating systems on your computer. This chapter  
describes the utility, and the configuration options it offers. Clustering is  
described at the end of the chapter.  
Configuration On Disk  
PERC 3 supports Configuration on Disk (drive roaming.) Configuration on  
Disk saves configuration information both in the PERC 3 non-volatile  
random access memory (NVRAM) and on the physical drives attached to  
the PERC 3. If the PERC 3 is replaced, the new PERC 3 controller can  
detect the RAID configuration, maintaining the integrity of the data on  
each drive even if the drives have changed target ID.  
Perform the following steps to enable configuration on disk support:  
1
Press <Ctrl> <M> at the PERC 3 power-on self test (POST) screen  
to run the BIOS Configuration Utility.  
2
3
4
5
6
Select Configure Menu—> View/Add Configuration.  
Select Disk when asked to use Disk or NVRAM.  
Select Save.  
Press <Esc> to exit the utility.  
Reboot the computer.  
Starting the PERC 3 BIOS  
Configuration Utility  
When the host computer boots, hold the <Ctrl> key and press the <M>  
key when the following appears:  
HA -0 (Bus 1 Dev 6) Type: PERC 3/QC Standard FW x.xx SDRAM=  
128MB  
0 Logical Drives found on the Host Adapter  
Adapter BIOS Disabled, No Logical Drives handled by BIOS  
0 Logical Drive(s) handled by BIOS  
150  
PERC 3 BIOS Configuration Utility  
Press <Ctrl><H> to Enable BIOS  
For each PERC adapter in the host system, the firmware version, dynamic  
random access memory (DRAM) size, and the status of logical drives on the  
card display. If you do not press <Ctrl><M> within a few seconds of the  
prompt, the computer continues normal booting.  
After you press a key to continue, the Management Menu screen displays.  
See the next section, “BIOS Configuration Utility Management Menu  
Options,” for more information.  
BIOS Configuration Utility Menu  
Options  
Table 11-1 describes the options for the BIOS Configuration Utility  
Management Menu. The menu and sub-menu options are explained in the  
following sections.  
Table 11-1. BIOS Configuration Utility Menu Options  
Option  
Description  
Configure  
Initialize  
Select this option to configure physical arrays and logical drives.  
Select this option to initialize one or more logical drives.  
PERC 3 BIOS Configuration Utility  
151  
Table 11-1. BIOS Configuration Utility Menu Options  
Option  
Description  
Objects  
Select this option to individually access controllers, logical drives,  
and physical drives.  
Format  
Rebuild  
Select this option to low-level format hard disk drives.  
Select this option to rebuild failed disk drives.  
Check  
Consistency  
Select this option to verify that the redundancy data in logical  
drives using RAID level 1, 5, 10, or 50 is correct.  
Select  
Adapter  
Select this option to list the adapters and select the adapter that  
you want to configure. The number of the selected adapter  
displays in the top right corner of the screen, along with its mode,  
either standard or cluster.  
Configure Menu  
Choose Configure to select a method for configuring arrays and logical  
drives. Table 11-2 displays the configuration methods, clear configuration  
option, and boot drive option.  
Table 11-2. Configuration Menu Options  
Option  
Description  
Easy  
Configuration  
Select this method to perform a logical drive configuration  
where every physical array you define is automatically  
associated with exactly one logical drive.  
New  
Configuration  
Select this method to discard the existing configuration  
information and to configure new arrays and logical drives. In  
addition to providing the basic logical drive configuration  
functions, New Configuration allows you to associate logical  
drives with multiple arrays (spanning.)  
View/Add  
Configuration  
Select this method to examine the existing configuration  
and/or to specify additional arrays and logical drives.  
View/Add Configuration provides the same functions  
available in New Configuration.  
Clear  
Configuration  
Select this option to erase the current configuration  
information from the PERC 3 non-volatile memory.  
Specify Boot  
Logical Drive  
Select this option to specify a logical drive as the boot drive  
on this adapter.  
152  
PERC 3 BIOS Configuration Utility  
Initialize Menu  
Choose Initialize from the BIOS Configuration Utility Management Menu  
to initialize one or more logical drives. Press the space bar to select a single  
driver or the <F2> key to select all drives for initialization. This action  
typically follows the configuration of a new logical drive.  
NOTE: See page 178 for  
procedures for initializing  
logical drives.  
NOTICE: Initializing a logical drive destroys all data on the logical drive.  
Objects Menu  
Choose Objects from the Management Menu to access the adapters,  
logical drives, physical drives, and SCSI channels individually. You can also  
change settings for each object. The Objects menu options are described in  
the following sections.  
Adapter  
Choose Objects—>Adapter to select a controller (if the computer has more  
than one) and to modify parameters. You can install only one PERC 3  
controller, but you can install other PERC controllers.  
Table 11-3 describes the Adapter menu options.  
Table 11-3. Adapter Menu Options  
Option  
Description  
Clear  
Configuration  
Choose this option to erase the current configuration from  
the controller non-volatile memory.  
FlexRAID  
PowerFail  
Choose this option to enable or disable the FlexRAID  
PowerFail feature. This option allows drive reconstruction,  
rebuild, and check consistency to continue when the system  
restarts because of a power failure, reset, or hard boot.  
Fast Initialization  
Select this option to write zeros to the first sector of the  
logical drive so that initialization occurs in 2 – 3 seconds.  
When this option is set to Disabled, a full initialization  
takes place on the entire logical drive. On a larger array  
(over 5 arrays), it is best to set fast initialization to  
Disabled, then initialize. Otherwise, the controller will run  
a background consistency check within five minutes of  
reboot or RAID 5 creation.  
PERC 3 BIOS Configuration Utility  
153  
Table 11-3. Adapter Menu Options (continued)  
Option  
Description  
Disk Spin up  
Timings  
Choose this option to set the method and timing for  
spinning up the hard disk drives.  
Cache Flush  
Timings  
Choose this option to set the cache flush interval to once  
every 2, 4, 6, 8, or 10 seconds. The default is 4.  
Rebuild Rate  
Use this option to select the rebuild rate for drives attached  
to the selected adapter.  
The rebuild rate is the percentage of the system resources  
dedicated to rebuilding a failed drive. A rebuild rate of 100  
percent means the system is totally dedicated to rebuilding  
the failed drive. The default is 30 percent.  
Alarm Control  
Choose this option to enable, disable, or silence the  
onboard alarm tone generator. The alarm sounds when  
there is a change in a drive state, such as when a drive fails  
or when a rebuild is complete.  
Other Adapter  
Information  
Provides general information about the adapter, such as the  
firmware version, and BIOS version.  
Factory Default  
Choose this option to load the default BIOS Configuration  
Utility settings.  
Enable BIOS  
Choose this option to enable or disable the BIOS on the  
adapter. If the boot device is on the RAID controller, the  
BIOS must be enabled; otherwise, the BIOS should be  
disabled or it might not be possible to use a boot device  
elsewhere.  
Emulation  
You can operate in the I2O mode or mass storage mode  
Dell recommends that you use only mass storage mode, and  
Dell drivers only.  
Auto Rebuild  
Initiator ID  
Set to Enabled to automatically rebuild drives when they  
fail.  
ID for the cluster card. It cannot have the same ID as the  
other node. The default is 7.  
Cluster Mode  
You can enable or disable cluster mode. When you disable  
cluster mode, the system operates in standard mode.  
Cluster:Adapteror Standard:Adapterdisplays in  
the upper right corner of the screen.  
154  
PERC 3 BIOS Configuration Utility  
Cluster Mode and the Initiator ID  
You can enable or disable cluster mode. When you disable cluster mode, the  
system operates in standard mode. In addition, when you enable cluster  
mode, the system automatically disables the BIOS.  
You can change the Initiator ID only when you are in cluster mode. You  
cannot change the ID while in standard mode. The ID can be a number  
from 0 to 15. We recommend that you use 6 or 7. When you are in standard  
mode, the ID is always 7.  
Logical Drive  
Choose Objects—> Logical Drive to select a logical drive and to perform  
the actions listed in Table 11-4.  
Table 11-4. Logical Drive Menu Options  
Option  
Description  
Initialize  
Initializes the selected logical drive. Do this for every logical  
drive that is configured.  
Check  
Consistency  
Verifies the correctness of the redundancy data in the selected  
logical drive. This option is available only if RAID level 1, 5, 10,  
or 50 is used. PERC 3 automatically corrects any differences  
found in the data.  
View/Update  
Parameters  
Displays the properties of the selected logical drive. You can  
modify the cache write policy, read policy, and the input/output  
(I/O) policy and can enable virtual sizing from this menu.  
Virtual Sizing  
Virtual sizing allows the PERC 3 controller to determine the drive capacity.  
The operating system reports the drive capacity as determined by PERC.  
Virtual sizing is not supported in cluster mode.  
Set virtual sizing to Enabled before you add a physical drive to a logical  
drive. Perform the following steps to enable virtual sizing.  
1
2
3
Select Objects—> Logical Drive.  
Select Logical Drive—>View/Update Properties.  
Select View/Update Properties—> Virtual Sizing.  
PERC 3 BIOS Configuration Utility  
155  
4
Select Enabled to enable Virtual Sizing.  
After you have created a logical drive set, the partition of the drive  
should be as large as the virtual size of the logical drive. After you have  
created a logical array set, the drive partition can be as large as the full  
size of the logical drive. However, this is the full virtual drive size, not  
the actual physical drive size.  
Physical Drive  
Select Objects—> Physical Drive to select a physical device and to perform  
the operations listed in the table below. The physical drives in the computer  
are listed. Move the cursor to the desired device and press <Enter> to  
display the screen.  
Table 11-5 displays the operations you can perform on the physical drives.  
Table 11-5. Physical Drive Menu Options  
Option  
Description  
Rebuild  
Rebuilds the selected physical drive.  
Low-level formats the selected disk drive.  
Changes the state of the selected disk drive to online.  
Changes the state of the selected disk drive to offline.  
Format  
Force Online  
Force Offline/  
Remove HSP  
Make HotSpare  
Designates the selected disk drive as a hot spare.  
View Drive  
Displays the drive properties for the selected physical device.  
Information  
View Rebuild  
Progress  
Indicates how much of the rebuild has been completed.  
SCSI Command Sets the number of queue tags per command. The options  
Qtagging  
are Disabled Command Tagging, 2 Queue Tags, 3 Queue  
Tags, 4 Queue Tags, or Enhanced QTag Scheduling. The  
default is Enhanced QTag Scheduling.  
Transfer Speed  
Option  
Selects the speed at which data is transferred. Displays a  
menu that contains the options Negotiation=Wide, and Set  
Transfer Speed. The maximum transfer speed is 160M.  
156  
PERC 3 BIOS Configuration Utility  
Channel  
Choose Objects—> Channel to select a SCSI channel on the currently  
selected controller. After you select a channel, press <Enter> to display the  
options for that channel. Table 11-6 describes the SCSI channel menu  
options.  
Table 11-6. SCSI Channel Menu Options  
Option  
Description  
Termination  
Enabled/Disabled  
When set to enabled, the PERC 3 controller is  
terminated. When set to disabled, it is not terminated.  
Normally, you do not need to change this setting; PERC 3  
automatically sets this option.  
SCSI Transfer Rate Used to select the SCSI transfer rate. The options are  
Fast, Ultra, Ultra-2, and 160M.  
NOTE: The disk transfer rate is set for each disk,  
while the SCSI channel transfer rate controls the  
speed of the bus. No matter how fast you set the disk  
transfer rate, the speed depends on the SCSI  
channel transfer rate.  
Battery Information  
Choose Objects—> Battery Information to view the battery backup  
information. Table 11-7 describes the battery information options.  
NOTE: The battery  
information is valid only  
for the PERC 3/QC and  
PERC 3/DC. The PERC  
3/DCL and PERC 3/SC do  
not have a battery option.  
Table 11-7. Batter y Information Menu Options  
Menu Item  
Explanation  
Backup Module  
PRESENT appears if there is a battery module; ABSENT if  
there is not.  
Battery Pack  
Temperature  
Voltage  
PRESENT appears if the battery pack is properly installed;  
ABSENT if it is not.  
GOOD appears if the temperature is within the normal  
range. HIGH appears if the module is too hot.  
GOOD appears if the voltage is within the normal range.  
BAD appears if the voltage is out of range.  
Fast Charging  
COMPLETED appears if the fast charge cycle is done.  
CHARGING appears if the battery pack is charging.  
PERC 3 BIOS Configuration Utility  
157  
Table 11-7. Batter y Information Menu Options (continued)  
Menu Item  
Explanation  
No of Cycles  
Displays the number of charge cycles. After 1100 charge  
cycles, the life of the battery pack is assumed to be over and  
you must replace it.  
Reset Battery Charge Counter  
Choose Objects—> Reset Battery Charge Counter to reset the charge cycle  
count. Perform the following steps to configure the charge cycle:  
1
2
3
When you first install a battery pack, set the Charge Cycle to 0.  
Select Reset Battery Charge Counter.  
Select YES to reset the number of cycles to zero and press <Enter>.  
After 1100 charge cycles, the life of the battery pack is assumed to be  
over and you must replace it.  
Format Menu  
Choose Format from the Management Menu to low-level format one or  
more physical drives.  
NOTICE: Formatting a hard drive destroys all data on the drive.  
Because most SCSI disk drives are low-level formatted at the factory, this  
step is usually not necessary. You must format a disk only if:  
•
•
The disk drive was not low-level formatted at the factory, or  
An excessive number of media errors have been detected on the disk  
drive.  
You do not have to choose Format to erase existing information on your  
SCSI disks, such as a system partition. That information is erased when you  
initialize the logical drive(s).  
Rebuild Menu  
Choose Rebuild from the Management Menu to rebuild one or more failed  
physical drives.  
158  
PERC 3 BIOS Configuration Utility  
Check Consistency Menu  
Choose Check Consistency to verify the redundancy data in logical drives  
that use RAID levels 1, 5, 10, and 50.  
The parameters of the existing logical drives appear. Discrepancies are  
automatically corrected, assuming always that the data is correct. However, if  
the failure is a read error on a data drive, the bad data block is reassigned  
with the generated data.  
Perform the following steps to run a consistency check:  
1
Select Check Consistency on the Management Menu.  
The currently configured logical drives display.  
2
3
Press the arrow keys to choose the desired logical drives.  
Press the spacebar to select or deselect a drive for consistency  
checking.  
4
5
Press <F2> to select or deselect all the logical drives.  
Press <F10> to begin the consistency check.  
A progress indicator for each selected logical drive displays.  
6
When the check is finished, press any key to clear the progress display.  
Press <Esc> to display the Management Menu.  
(To check an individual drive, you can select Objects—> Logical  
Drives, the desired logical drive(s), and Check Consistency on the  
action menu.)  
Configuring Arrays and Logical Drives  
You can configure physical arrays and logical drives with the BIOS  
Configuration Utility using:  
•
•
•
Easy Configuration  
New Configuration  
View/Add Configuration  
Each configuration method requires a different level of user input. The  
general flow of operations for array and logical drive configuration is:  
PERC 3 BIOS Configuration Utility  
159  
1
2
3
4
5
6
Choose a configuration method.  
Designate hot spares (optional).  
Create arrays using the available physical drives.  
Define logical drives using the space in the arrays.  
Save the configuration information.  
Initialize the new logical drives.  
Choosing the Configuration Method  
Easy Configuration  
In Easy Configuration, each physical array you create is associated with  
exactly one logical drive, and you can modify the following parameters:  
•
•
•
•
•
RAID level  
Stripe size  
Write policy  
Read policy  
I/O policy  
If logical drives have already been configured when you select Easy  
Configuration, the configuration information is not disturbed.  
New Configuration  
In New Configuration, you can modify the following logical drive  
parameters:  
•
•
•
•
•
•
•
RAID level  
Stripe size  
Write policy  
Read policy  
I/O policy  
Logical drive size  
Spanning of arrays  
160  
PERC 3 BIOS Configuration Utility  
If you select New Configuration, the existing configuration information on  
the selected controller is destroyed when the new configuration is saved.  
View/Add Configuration  
View/Add Configuration allows you to control the same logical drive  
parameters as New Configuration without disturbing the existing  
configuration information. You can also choose to enable the Configuration  
on Disk feature.  
PERC 3/QC and PERC 3/DC Default Settings  
Table 11-8 displays the default settings for PERC 3/QC and PERC 3/DC.  
NOTE: In cluster mode,  
PERC 3/DC allows write-  
through cache policy only.  
Table 11-8. PERC 3/QC and PERC 3/DC Default  
Settings  
Stripe size  
64 KB  
Write policy  
Write-back  
Read-ahead policy  
Cache policy  
Adaptive  
Direct I/O  
FlexRAID virtual sizing  
FlexRAID PowerFail  
OFF  
ON (OFF for cluster enabled)  
PERC 3/SC and PERC 3/DCL Default Settings  
Table 11-9 displays the default settings for PERC 3/SC and PERC 3/DCL.  
Table 11-9. PERC 3/SC and PERC 3/DCL Default  
Settings  
Stripe size  
64 KB  
Write policy  
Write-through  
Adaptive  
Direct I/O  
OFF  
Read-ahead policy  
Cache policy  
FlexRAID virtual sizing  
FlexRAID PowerFail  
ON  
PERC 3 BIOS Configuration Utility  
161  
Reserved Disk Space during Configuration  
Up to 20.6 MB of disk space is reserved when a hard drive is being  
configured.  
Designating Drives as Hot Spares  
Hot spares are physical drives that are powered up along with the RAID  
drives and usually stay in a standby state. The hot spare drive should be  
larger than or equal to the smallest logical drive. If a hard drive used in a  
RAID logical drive fails, a hot spare will automatically take its place and the  
data on the failed drive is reconstructed on the hot spare. Hot spares can be  
used for RAID levels 1 and 5. Each PERC 3 controller supports up to eight  
hot spares.  
The methods for designating physical drives as hot spares are:  
•
Pressing <F4> while creating arrays in Easy, New or View/Add  
Configuration mode, or  
•
Using the Objects—> Physical Drive menu.  
Press <F4>  
When you choose any configuration option, a list of all physical devices  
connected to the current controller appears. Perform the following steps to  
designate a drive as a hot spare.  
1
2
Press the arrow keys to choose a disk drive that has a READY indicator  
Press <F4> to designate the drive as a hot spare.  
The indicator will change to HOTSP.  
3
Save the configuration.  
Objects Menu  
Perform the following steps to designate a drive as a hot spare.  
1
2
Select Objects—> Physical Drive from the Management Menu.  
A physical drive selection screen displays.  
Select a hard drive and press <Enter> to display the action menu for  
the drive.  
162  
PERC 3 BIOS Configuration Utility  
3
4
Press the arrow keys to select Make HotSpare and press <Enter>.  
The indicator for the selected drive changes to HOTSP.  
Save the configuration.  
Using Easy Configuration  
In Easy Configuration, each array is associated with exactly one logical  
drive. Perform the following steps to create arrays using Easy  
Configuration.  
1
Choose Configure from the Management Menu.  
The Configure options menu displays.  
2
Choose Configure—> Easy Configuration.  
The array selection menu appears. Hot key information displays at the  
bottom of the screen. The hot key functions are:  
–
<F2> Display the manufacturer data and error count for the  
selected drive.  
–
–
<F3> Display the logical drives that have been configured.  
<F4> Designate the selected drive as a hot spare.  
PERC 3 BIOS Configuration Utility  
163  
NOTE: Try to use drives  
of the same capacity in a  
specific array. If you use  
drives with different  
3
4
Press the arrow keys to highlight specific physical drives.  
Press the spacebar to associate the selected physical drive with the  
current array.  
capacities in an array, all  
drives in the array are  
treated as if they have the  
capacity of the smallest  
drive in the array.  
The indicator for the selected drive changes from READY to ONLIN  
A[array number]-[drive number]. For example, ONLIN A2-3 means  
disk drive 3 in array 2.  
5
Add physical drives to the current array as desired.  
The number of physical drives in a specific array determines the RAID  
levels that can be implemented with the array.  
–
–
–
–
–
RAID 0 requires one or more physical drives.  
RAID 1 requires exactly two physical drives.  
RAID 5 requires at least three physical drives.  
RAID 10 requires at least four physical drives.  
RAID 50 requires at least six physical drives.  
6
Press <Enter> after you finish creating the current array.  
A window entitled Select Configurable Array(s) appears. It displays  
the array, and array number, such as A-00.  
7
Press the spacebar to select the array.  
Span information, such as Span-1, displays in the array box. You can  
create multiple arrays, then select them to span them.  
164  
PERC 3 BIOS Configuration Utility  
NOTE: You can press  
<F2> to display the  
number of drives in the  
array, their channel and  
ID, and <F3> to display  
array information, such  
as the stripes, slots, and  
free space.  
8
Press <F10> to configure logical drives.  
The logical drive configuration screen appears. Span=Yes displays on  
this screen if you select two or more arrays to span.  
The window from the top of the screen shows the logical drive that is  
currently being configured as well as any existing logical drives.  
The column headings are:  
–
–
–
–
LD: The logical drive number  
RAID: The RAID level  
Size: The logical drive size  
#Stripes: The number of stripes (physical drives) in the  
associated physical array  
–
–
StrpSz: The stripe size  
DriveState: The state of the logical drive  
9
Highlight RAID and press <Enter>.  
The available RAID levels for the current logical drive display.  
10 Select a RAID level for the logical drive and press <Enter> to  
confirm.  
See Chapter 3 for an explanation of the RAID levels.  
11 Set the Stripe Size from the Advanced Menu.  
PERC 3 BIOS Configuration Utility  
165  
Stripe Size specifies the size of the segments written to each disk in a  
RAID 1 or 5 logical drive. You can set the stripe size to 2 KB, 4 KB,  
8 KB, 16 KB, 32 KB, 64 KB, or 128 KB. A larger stripe size produces  
better read performance, especially if your computer does mostly  
sequential reads. If you are sure that your computer does random read  
requests more often, choose a small stripe size. The default is 64 KB.  
12 Set the Write Policy from the Advanced Menu.  
Write Policy sets the caching method to write-back or write-through.  
NOTE: In Cluster mode,  
PERC 3/DC allows write-  
through cache policy only.  
–
In Write-back caching, the controller sends a data transfer  
completion signal to the host when the controller cache has  
received all the data in a transaction. This setting is recommended  
in standard mode.  
–
In Write-through caching, the controller sends a data transfer  
completion signal to the host when the disk subsystem has  
received all the data in a transaction. This is the default setting if  
cluster mode is enabled.  
Write-through caching has a data security advantage over write-back  
caching. Write-back caching has a performance advantage over write-  
through caching.  
13 Set the Read Policy from the Advanced Menu.  
The Read-ahead option enables the SCSI read-ahead feature for the  
logical drive. You can set this parameter to No-Read-Ahead, Read-  
ahead, or Adaptive. Adaptive is the default setting.  
–
–
–
No-Read-Ahead specifies that the controller does not use read-  
ahead for the current logical drive.  
Read-ahead specifies that the controller uses read-ahead for the  
current logical drive.  
Adaptive specifies that the controller begins using read-ahead if  
the two most recent disk accesses occurred in sequential sectors. If  
all read requests are random, the algorithm reverts to No-Read-  
Ahead, however, all requests are still evaluated for possible  
sequential operation.  
14 Set the Read Policy from the Advanced Menu.  
Cache Policy applies to reads on a specific logical drive. It does not  
affect the Read-ahead cache.  
–
Cached I/O specifies that all reads are buffered in cache memory.  
166  
PERC 3 BIOS Configuration Utility  
–
Direct I/O specifies that reads are not buffered in cache memory.  
Direct I/O does not override the cache policy settings. Data is  
transferred to cache and the host concurrently. If the same data  
block is read again, it comes from cache memory. This is the  
default setting.  
15 Press <Esc> to exit the Advanced Menu.  
16 After you define the current logical drive, choose Accept and press  
NOTE: The PERC 3  
family supports spanning  
across RAID 1 and 5  
arrays only.  
<Enter>.  
The array selection screen appears if any unconfigured disk drives  
remain.  
17 Repeat steps 3 through 16 to configure another array and logical drive.  
PERC 3 supports up to forty logical drives per controller.  
18 If you are finished configuring logical drives, press <Esc> to exit Easy  
Configuration.  
A list of the currently configured logical drives appears.  
19 Save the configuration at the prompt.  
After you respond to the Save prompt, the Configure menu appears.  
20 Initialize the logical drives you have just configured.  
Using New Configuration  
The New Configuration option allows you to associate logical drives with  
multiple physical arrays (spanning of arrays.)  
NOTICE: Choosing the New Configuration option erases the existing  
configuration information on the selected controller. To use the spanning  
feature and keep the existing configuration, use View/Add  
Configuration.  
Perform the following steps to configure the arrays using New  
Configuration.  
1
Choose Configure—> New Configuration from the Management  
Menu.  
PERC 3 BIOS Configuration Utility  
167  
An array selection window displays the devices connected to the  
current controller. Hot key information appears at the bottom of the  
screen. The hot key functions are:  
–
<F2> Display drive data and PERC 3 error count for the selected  
drive.  
–
–
–
<F3> Display the logical drives that have been configured.  
<F4> Designate the selected drive as a hot spare.  
<F10> Display the logical drive configuration screen.  
NOTE: Try to use drives  
of the same capacity in a  
specific array. If you use  
drives with different  
2
3
Press the arrow keys to choose specific physical drives.  
Press the spacebar to associate the selected physical drive with the  
current array.  
capacities in an array, all  
drives in the array are  
treated as if they have the  
capacity of the smallest  
drive in the array.  
The indicator for the selected drive changes from READY to  
ONLINE A[array number]-[drive number]. For example, ONLINE  
A2-3 means disk drive 3 in array 2.  
4
Add physical drives to the current array as desired.  
The number of physical drives in a specific array determines the RAID  
levels that can be implemented with the array.  
–
–
–
–
RAID 0 requires one or more physical drives per array.  
RAID 1 requires two physical drives per array.  
RAID 5 requires at least three physical drives per array.  
RAID 10 requires at least four physical drives per array.  
168  
PERC 3 BIOS Configuration Utility  
–
RAID 50 requires at least six physical drives per array.  
5
Press <Enter> after you finish creating the current array.  
A window entitled Select Configurable Array(s) appears. It displays  
the array, and array number, such as A-00.  
6
Press the spacebar to select the array.  
Span information, such as Span-1, displays in the array box. You can  
create multiple arrays, then select them to span them.  
NOTE: You can press  
<F2> to display the  
number of drives in the  
array, their channel and  
ID, and <F3> to display  
array information, such  
as the stripes, slots, and  
free space.  
7
Press <F10> to configure a logical drive.  
PERC 3 BIOS Configuration Utility  
169  
The logical drive configuration screen appears. Span=Yes displays on  
this screen if you select two or more arrays to span.  
The window at the top of the screen shows the logical drive that is  
currently being configured as well as any existing logical drives. The  
column headings are:  
–
–
–
–
LD: The logical drive number  
RAID: The RAID level  
Size: The logical drive size  
#Stripes: The number of stripes (physical drives) in the  
associated physical array  
–
–
StrpSz: The stripe size  
Drive-State: The state of the logical drive  
8
9
Highlight RAID and press <Enter>.  
A list of the available RAID levels for the current logical drive appears.  
Select a RAID level for the logical drive and press <Enter> to  
confirm.  
See Chapter 3 for an explanation of the RAID levels.  
10 Set the spanning mode for the current logical drive.  
11 Highlight Span and press <Enter>.  
The choices are:  
–
CanSpan: Array spanning is enabled for the current logical drive.  
The logical drive can occupy space in more than one array.  
–
NoSpan: Array spanning is disabled for the current logical drive.  
The logical drive can occupy space in only one array.  
PERC 3 supports spanning of RAID 1 and 5 arrays only. You can span  
two or more contiguous RAID 1 logical drives into a RAID 10 array,  
and two or more contiguous RAID 5 logical drives into a RAID 50  
array.  
For two arrays to be spanned, they must have the same stripe width  
(they must contain the same number of physical drives) and must be  
consecutively numbered. For example, assuming Array 2 contains four  
disk drives, it can be spanned only with Array 1 and/or Array 3, and  
only if Arrays 1 and 3 also contain four physical drives. If the two  
criteria for spanning are met, PERC 3 automatically allows spanning. If  
170  
PERC 3 BIOS Configuration Utility  
the criteria are not met, the span setting makes no difference for the  
current logical drive.  
12 Highlight a spanning option and press <Enter>.  
13 Set the logical drive size.  
14 Move the cursor to Size and press <Enter>.  
By default, the logical drive size is set to all available space in the  
array(s) being associated with the current logical drive, accounting for  
the span setting.  
15 Open the Advanced Menu to set the remaining options:  
16 Set the Stripe Size.  
Stripe size specifies the size of the segments written to each disk in a  
RAID 1 or 5 logical drive. You can set the stripe size to 2 KB, 4 KB, 8  
KB, 16 KB, 32 KB, 64 KB, or 128 KB. A larger stripe size produces  
higher read performance, especially if your computer does mostly  
sequential reads. However, if you are sure that your computer does  
random read requests more often, select a small stripe size. The  
default stripe size is 64 KB.  
17 Set the Write Policy.  
Write Policy sets the caching method to write-back or write-through.  
–
In Write-back caching, the controller sends a data transfer  
completion signal to the host when the controller cache has  
received all the data in a transaction. This setting is recommended  
in standard mode.  
–
In Write-through caching, the controller sends a data transfer  
completion signal to the host when the disk subsystem has  
received all the data in a transaction. This is the default setting in  
cluster mode.  
Write-through caching has a data security advantage over write-back  
caching, whereas write-back caching has a performance advantage over  
write-through caching.  
18 Set the Read Policy.  
Read-ahead This option enables the SCSI read-ahead feature for the  
logical drive. You can set this parameter to No-Read-Ahead, Read-  
ahead, or Adaptive. The default setting is Adaptive.  
PERC 3 BIOS Configuration Utility  
171  
–
–
–
No-Read-Ahead specifies that the controller does not use read-  
ahead for the current logical drive.  
Read-ahead specifies that the controller uses read-ahead for the  
current logical drive.  
Adaptive specifies that the controller begins using read-ahead if  
the two most recent disk accesses occurred in sequential sectors. If  
all read requests are random, the algorithm reverts to No-Read-  
Ahead, however, all requests are still evaluated for possible  
sequential operation. This is the default setting.  
19 Set the Cache Policy.  
Cache Policy applies to reads on a specific logical drive. It does not  
affect the read ahead cache.  
–
–
Cached I/O specifies that all reads are buffered in cache memory.  
Direct I/O specifies that reads are not buffered in cache memory.  
This is the default setting. Direct I/O does not override the cache  
policy settings. Data is transferred to cache and the host  
concurrently. If the same data block is read again, it comes from  
cache memory.  
20 Press <Esc> to exit the Advanced Menu.  
21 After you define the current logical drive, choose Accept and press  
<Enter>.  
If space remains in the arrays, the next logical drive to be configured  
appears.  
22 Repeat steps 3 to 22 to create an array and configure another logical  
drive.  
If the array space has been used, a list of the existing logical drives  
appears.  
23 Press any key to continue and respond to the Save prompt.  
24 Initialize the logical drives you have just configured.  
172  
PERC 3 BIOS Configuration Utility  
Using View/Add Configuration  
View/Add Configuration allows you to associate logical drives with physical  
arrays (this is called spanning of arrays.) The existing configuration is left  
intact, so you can also use View/Add Configuration simply to look at the  
current configuration. Perform the following steps to configure the arrays  
using View/Add Configuration.  
1
Choose Configure from the Management Menu.  
The Configure menu options display.  
2
Choose Configure—> View/Add Configuration.  
An array selection window displays the devices connected to the  
current controller.  
Hot key information appears at the bottom of the screen. The hot key  
functions are:  
–
<F2> Display the manufacturer data and PERC 3 error count for  
the selected drive.  
–
–
–
<F3> Display the logical drives that have been configured.  
<F4> Designate the selected drive as a hot spare.  
<F10> Display the logical drive configuration screen.  
PERC 3 BIOS Configuration Utility  
173  
NOTE: Try to use drives  
of the same capacity in a  
specific array. If you use  
drives with different  
3
4
Press the arrow keys to highlight specific physical drives.  
Press the spacebar to associate the selected physical drive with the  
current array.  
capacities in an array, all  
drives in the array are  
treated as if they have the  
capacity of the smallest  
drive in the array.  
The indicator for the selected drive changes from READY to ONLIN  
A[array number]-[drive number]. For example, ONLIN A2-3 means  
physical drive 3 in array 2.  
5
Add physical drives to the current array as desired.  
The number of physical drives in a specific array determine the RAID  
levels that can be implemented with the array.  
NOTE: Dell does not  
recommend using RAID 0  
in the cluster  
environment. RAID 0 is  
not redundant and  
provides no data  
–
–
–
–
–
RAID 0 requires one or more physical drives per array.  
RAID 1 requires two physical drives per array.  
RAID 5 requires at least three physical drives per array.  
RAID 10 requires at least four physical drives per array.  
RAID 50 requires at least six physical drives per array.  
protection.  
6
Press <Enter> after you finish creating the current array.  
A window entitled Select Configurable Array(s) appears. It displays  
the array, and array number, such as A-00.  
7
Press the spacebar to select the array.  
Span information, such as Span-1, displays in the array box. You can  
create multiple arrays, then select them to span them.  
174  
PERC 3 BIOS Configuration Utility  
NOTE: You can press  
<F2> to display the  
number of drives in the  
array, their channel and  
ID, and <F3> to display  
array information, such  
as the stripes, slots, and  
free space.  
8
Press <F10> to configure a logical drive.  
The logical drive configuration screen appears. Span=Yes displays on  
this screen if you select two or more arrays to span.  
The logical drive that is currently being configured and any existing  
logical drives are displayed. The column headings are:  
–
–
–
–
LD: The logical drive number  
RAID: The RAID level  
Size: The logical drive size  
#Stripes: The number of stripes (physical drives) in the  
associated physical array  
–
–
StrpSz: The stripe size  
Drive-State: The state of the logical drive  
9
Highlight RAID and press <Enter>.  
The available RAID levels for the current logical drive appear.  
10 Select a RAID level for the logical drive and press <Enter> to  
confirm.  
See Chapter 3 for an explanation of the RAID levels.  
11 Set the spanning mode for the current logical drive. Highlight Span  
and press <Enter>.  
The choices are:  
PERC 3 BIOS Configuration Utility  
175  
–
–
CanSpan: Array spanning is enabled for the current logical drive.  
The logical drive can occupy space in more than one array.  
NoSpan: Array spanning is disabled for the current logical drive.  
The logical drive can occupy space in only one array.  
PERC 3 supports spanning for RAID 1 and RAID 5 arrays only. You can  
span two or more contiguous RAID 1 logical drives into a RAID 10  
array, or two or more contiguous RAID 5 logical drives into a RAID 50  
array.  
For two arrays to be spanned, they must have the same stripe width  
(they must contain the same number of physical drives) and they must  
be consecutively numbered. For example, assuming Array 2 contains  
four disk drives, it can be spanned only with Array 1 and/or Array 3,  
and only if Arrays 1 and 3 also contain four physical drives.  
If the two criteria for spanning are met, PERC 3 automatically  
activates spanning. If the criteria are not met, the span setting makes  
no difference for the current logical drive.  
12 Highlight a spanning option and press <Enter>.  
13 Set the logical drive size.  
14 Move the cursor to Size and press <Enter>.  
By default, the logical drive size is set to all available space in the  
array(s) associated with the current logical drive, accounting for the  
span setting.  
15 Open the Advanced Menu to set the remaining options.  
16 Set the Stripe Size.  
Stripe size specifies the size of the segment written to each disk in a  
RAID 1 or 5 logical drive. You can set the stripe size to 2 KB, 4 KB, 8  
KB, 16 KB, 32 KB, 64 KB, or 128 KB. A larger stripe size produces  
higher read performance, especially if your computer does mostly  
sequential reads. However, if your computer does random read  
requests more often, choose a smaller stripe size. The default is 64 KB.  
17 Set the Write Policy.  
Write Policy specifies the cache write policy. You can set the write  
policy to Write-back or Write-through:  
176  
PERC 3 BIOS Configuration Utility  
–
–
In Write-back caching, the controller sends a data transfer  
completion signal to the host when the controller cache has  
received all the data in a transaction. This setting is not  
recommended.  
In Write-through caching, the controller sends a data transfer  
completion signal to the host when the disk subsystem has  
received all the data in a transaction. This is the default setting for  
cluster mode.  
Write-through caching has a data security advantage over write-back  
caching, whereas write-back caching has a performance advantage over  
write-through caching.  
18 Set the Read Policy.  
Read-ahead enables the SCSI read-ahead feature for the logical drive.  
You can set this parameter to No-Read-Ahead, Read-ahead, or  
Adaptive. Adaptive is the default setting.  
–
–
–
No-Read-Ahead specifies that the controller does not use read-  
ahead for the current logical drive.  
Read-ahead specifies that the controller uses read-ahead for the  
current logical drive.  
Adaptive specifies that the controller begins using read-ahead if  
the two most recent disk accesses occurred in sequential sectors. If  
all read requests are random, the algorithm reverts to No-Read-  
Ahead, however, all requests are still evaluated for possible  
sequential operation.  
19 Set the Read Policy.  
Cache Policy applies to reads on a specific logical drive. It does not  
affect the Read-ahead cache.  
–
–
Cached I/O specifies that all reads are buffered in cache memory.  
Direct I/O specifies that reads are not buffered in cache memory.  
This is the default setting.  
Direct I/O does not override the cache policy settings. Data is  
transferred to cache and the host concurrently. If the same data block  
is read again, it comes from cache memory.  
20 Press <Esc> to exit the Advanced Menu.  
21 After you define the current logical drive, choose Accept and press  
<Enter>.  
PERC 3 BIOS Configuration Utility  
177  
If space remains in the arrays, the next logical drive to be configured  
appears.  
22 Repeat steps 3 to 21 to create an array and configure another logical  
drive.  
If all array space is used, a list of the existing logical drives appears.  
23 Press any key to continue.  
24 Respond to the Save prompt.  
25 Initialize the logical drives you have just configured.  
Initializing Logical Drives  
Initialize each new logical drive you configure. You can initialize the logical  
drives using:  
•
Batch Initialization. The Initialize option on the Management Menu  
lets you initialize up to eight logical drives simultaneously.  
•
Individual Initialization. The Objects—> Logical Drive action menu  
for an individual logical drive has an Initialize option.  
Batch Initialization  
To initialize logical drives using the batch initialization procedure:  
1
Choose Initialize from the BIOS Configuration Utility Management  
Menu.  
A list of the current logical drives appears.  
2
Press the spacebar to select the desired logical drives for initialization  
or press <F2> to select/deselect all logical drives.  
3
4
After you select the logical drives to initialize, press <F10>.  
Select Yes from the confirmation prompt.  
The progress of the initialization for each drive displays in bar graph  
format.  
5
6
When initialization is complete, press any key to continue.  
Press <Esc> to display the Management Menu.  
178  
PERC 3 BIOS Configuration Utility  
Individual Initialization  
1
Choose Objects—> Logical Drive option from the Management  
Menu.  
2
3
Select the logical drive to be initialized.  
Choose Initialize from the action menu.  
Initialization progress appears as a bar graph on the screen.  
4
When initialization completes, press any key to display the previous  
menu.  
Deleting Logical Drives (Random Array  
Deletion)  
This RAID controller supports random array deletion. Random array  
deletion is the ability to delete any unwanted logical drives and use that  
space for a new logical drive.  
The main benefit of random array deletion on the configuration module is  
that you are not restricted to sequential or contiguous logical drives when  
you create logical drives. You can use non-contiguous segments to create  
logical drives.  
NOTE: For more  
information about  
random array deletion,  
refer to page 99.  
After you delete a logical drive, you can create a new one. You can use the  
configuration utilities to create the next logical drive from the non-  
contiguous free space (‘holes’), and from the newly created arrays. The  
configuration utility provides a list of configurable arrays where there is a  
space to configure.  
NOTICE: The deletion of the logical drive can fail under certain conditions.  
Deletion can fail during a rebuild, initialization or check consistency of a  
logical drive, if that drive has a higher logical drive number than the drive you  
want to delete.  
Perform the following steps to delete logical drives:  
1
Choose Objects—> Logical Drive from the Management Menu.  
The logical drives display.  
2
3
Use the arrow key to highlight the logical drive you want to delete.  
Press <F5> to delete the logical drive.  
PERC 3 BIOS Configuration Utility  
179  
This deletes the logical drive and makes the space it occupied available  
for you to make another logical drive.  
Formatting Physical Drives  
You can do low-level formatting of SCSI drives using the BIOS  
Configuration Utility. Because most SCSI disk drives are low-level  
formatted at the factory, this step is usually not necessary. Usually, you must  
format a disk if:  
•
•
The disk drive was not low-level formatted at the factory, or  
An excessive number of media errors have been detected on the disk  
drive.  
Media Errors  
Perform the following steps to view the media errors for the drive to be  
formatted.  
1
2
3
Select Objects—> Physical Drives from the Management Menu.  
Select the Physical Drives option, and choose a device.  
Press <F2>.  
The View Drive Information screen displays. The error count displays  
at the bottom of the properties screen. If you feel that the number of  
errors is excessive, you should probably format the disk drive. If more  
than 32 media errors were detected, PERC 3 automatically puts the  
drive in FAIL state. This occurs even in a degraded RAID set. The  
errors display as they occur. When there are errors, formatting the  
drive can clear up the problem.  
You do not have to choose Format to erase existing information on  
your SCSI disks, such as a DOS partition. That information is erased  
when you initialize logical drives.  
Formatting Drives  
You can format the physical drives using:  
•
Batch Formatting. The Format option in the PERC 3 BIOS  
Configuration Utility main menu lets you format up to eight disk  
drives simultaneously.  
180  
PERC 3 BIOS Configuration Utility  
•
Individual Formatting. Choose the Format option from Objects—>  
Physical Drive action menu for a physical drive.  
Batch Formatting  
1
Select Format from the Management Menu.  
A device selection window displays the devices connected to the  
current controller.  
2
3
Press the arrow keys to highlight the drives.  
Press the spacebar to select the desired physical drives for formatting.  
The indicators for the selected drives flashes.  
4
After you select the physical drives, press <F10> and choose Yes from  
the confirmation prompt.  
The indicators for the selected drives changes to FRMT[number],  
where number reflects the order of drive selection. Formatting may  
take some time, depending on the number of drives you have selected  
and the drive capacities.  
5
6
When formatting is complete, press any key to continue.  
Press <Esc> to display the Management Menu.  
Individual Formatting  
1
Choose Objects—> Physical Drive from the Management Menu.  
A device selection window displays the devices connected to the  
current controller.  
2
3
Press the arrow keys to select the physical drive to be formatted, then  
press <Enter>.  
Choose the Format option.  
Formatting can take some time, depending on the drive capacity.  
4
When formatting completes, press any key to display the previous  
menu.  
PERC 3 BIOS Configuration Utility  
181  
Rebuilding Failed Disk Drives  
If a disk drive fails in an array that is configured as a RAID 1 or 5 logical  
drive, you can recover the lost data by rebuilding the drive. Table 11-10  
describes the types of rebuilds.  
Table 11-10. Rebuild Types  
Type  
Description  
Automatic If you have configured hot spares, PERC 3 automatically tries to  
Rebuild  
use them to rebuild failed disks. Display the Objects—> Physical  
Drive screen while a rebuild is in progress. The drive indicator for  
the hot spare disk drive has changed to REBLD A[array number]-  
[drive number], indicating the disk drive being replaced by the hot  
spare.  
Manual  
Rebuild  
Manual rebuild is necessary if no hot spares with enough capacity to  
rebuild the failed drives are available. Select Rebuild on the  
Management Menu or Rebuild on the Objects—> Physical Drive  
menu.  
Manual Rebuild – Rebuilding an Individual Drive  
Perform the following steps to rebuild an individual drive.  
1
Choose Objects—> Physical Drive from the Management Menu.  
A device selection window displays the devices connected to the  
current controller.  
2
3
Press the arrow keys to select the physical drive to rebuild, then press  
<Enter>.  
Choose Rebuild from the action menu and respond to the  
confirmation prompt.  
Rebuilding can take some time, depending on the drive capacity.  
When rebuild completes, press any key to display the previous menu.  
4
Manual Rebuild – Batch Mode  
Perform the following steps to rebuild drives in batch mode.  
1
Choose Rebuild from the Management Menu.  
182  
PERC 3 BIOS Configuration Utility  
A device selection window displays the devices connected to the  
current controller. The failed drives have FAIL indicators.  
2
3
4
5
Press the arrow keys to highlight any failed drives to be rebuilt.,  
Press the spacebar to select the desired physical drives for rebuild.  
After you select the physical drives, press <F10>.  
Select Yes at the prompt.  
The indicators for the selected drives change to REBLD. Rebuilding  
can take some time, depending on the number of drives selected and  
the drive capacities.  
6
7
When rebuild is complete, press any key to continue.  
Press <Esc> to display the Management Menu.  
Using a Pre-loaded SCSI Drive “As-is”  
You may have a SCSI hard drive that is already loaded with software. The  
drive may be a boot disk containing an operating system. If so, add the  
PERC device driver to this system drive before you switch to the PERC 3  
and attempt to boot from it. You can use the PERC 3 controller as a SCSI  
adapter for such a drive by performing the following steps.  
NOTE: To use a pre-  
loaded system drive in the  
manner described here,  
you must make it the first  
logical drive defined (for  
example: LD1) on the  
controller it is connected  
to. This will make the  
drive ID 0 LUN 0. If the  
drive is not a boot device,  
the logical drive number  
is not critical.  
1
Connect the SCSI drive to the channel on the PERC 3 controller, with  
proper termination and TID settings.  
2
3
4
5
6
Boot the computer.  
Press <Ctrl><M> to start the BIOS Configuration Utility.  
Select Configure—> Easy Configuration.  
Press the cursor keys to select the pre-loaded drive.  
Press the spacebar.  
The pre-loaded drive should now become an array element.  
Press <Enter>.  
7
You have now declared the pre-loaded drive as a one-disk array. The  
logical drive configuration screen displays.  
8
9
Set the Read Policy on the Advanced Menu.  
Set the Cache Policy on the Advanced Menu.  
PERC 3 BIOS Configuration Utility  
183  
10 Press <Esc> to exit the Advanced Menu.  
You are prompted to accept the settings.  
11 Highlight Accept and press <Enter>.  
Do not initialize the logical drive.  
12 Press <Esc>. and choose Yes at the Save prompt.  
13 Exit the BIOS Configuration Utility.  
14 Reboot the system.  
15 Set the host system to boot from SCSI, if such a setting is available.  
Exiting the PERC 3 BIOS  
Configuration Utility  
Perform the following steps to exit the BIOS Configuration Utility.  
1
2
3
Press <Esc> when the Management Menu displays.  
Choose Yes at the prompt.  
Reboot the computer.  
The PERC 3 BIOS message appears again.  
Press <Esc> when the BIOS Configuration Utility prompt appears.  
4
Setting Hardware Termination  
If you are using the PERC 3/DC for clustering, you must use hardware  
termination. Otherwise, software termination may be used. The headers  
used for SCSI termination are:  
•
•
•
•
J2 Termination Enable is a three-pin header that specifies control of  
the SCSI termination for channel 0.  
J3 Termination Enable is a three-pin header that specifies control of  
the SCSI termination for channel 1.  
J5 Termination Enable is a three-pin header that specifies control of  
the SCSI termination for channel 2.  
J7 Termination Enable is a three-pin header that specifies control of  
the SCSI termination for channel 3.  
184  
PERC 3 BIOS Configuration Utility  
To enable hardware termination, leave the pins open. The default is  
hardware termination.  
Clustering  
Physically, a cluster is a grouping of two independent servers that can access  
the same data storage and provide services to a common set of clients. With  
current technology, this usually means servers connected to common I/O  
buses and a common network for client access.  
NOTE: Clustering is not  
supported by PERC 3/QC,  
PERC 3/DCL or PERC  
3/SC. It is supported by  
PERC 3/DC.  
Logically, a cluster is a single management unit. Any server can provide any  
available service to any authorized client. The servers must have access to  
the same data and must share a common security model. Again, this  
generally means that the servers in a cluster will have the same architecture  
and run the same version of the same operating system.  
Cluster Mode  
You can use the PERC BIOS Configuration Utility to enable or disable  
cluster mode. When you enable cluster mode, the system operates in  
cluster mode. When you disable cluster mode, the system operates in  
standard mode. The following section explains how to do this.  
Enabling and Disabling the Cluster Mode  
NOTE: You can ignore  
this section on enabling  
and disabling cluster  
mode if you do not use  
clustering.  
You can enable or disable cluster mode. When you enable cluster mode, the  
system automatically disables the BIOS. If any of the controllers are set for  
clustering, the following warning appears when you press <Ctrl> <M> to  
access the BIOS Configuration Utility:  
!!!STOP!!!  
This operation may change the configuration of disks and  
can cause loss of data! Ensure:  
1. Peer server is powered up for its controller NVRAM to be  
updated. Otherwise, disk configuration should be read from  
disk and saved to the controller’s NVRAM.  
2. The second server must not be configuring the disks.  
3. There is no I/O activity against shared drives.  
4. Press Any Key to Continue.  
PERC 3 BIOS Configuration Utility  
185  
After you start the BIOS configuration utility, perform the following steps to  
enable or disable the cluster mode:  
1
2
Select Objects—> Adapter—> Cluster Mode.  
Select Enable or Disable.  
After you select Disable or Enable, a dialog box displays for you to  
confirm your choice to change the parameter. If Cluster Mode is  
enabled and you select Disable, a warning against connecting to  
shared cluster storage appears, and the dialog box asks whether you  
want to continue.  
If you select YES to disable, Cluster Mode is disabled, and the  
Initiator ID is set to 7.  
3
4
Reboot your system after the cluster mode is disabled.  
To change Cluster Mode from disabled to enabled, select Cluster  
Mode.  
A dialog box displays for you to choose to enable Cluster Mode.  
5
Select Enable.  
After you select Enable, a dialog box displays for you to confirm your  
choice to change the parameter.  
6
7
Select Yes to confirm.  
Cluster Mode changes to enabled.  
After you enable Cluster Mode, move the cursor to Initiator ID to  
change the Initiator ID.  
8
9
Press <Enter>.  
At the Change Initiator ID dialog box, select Yes.  
You can change the Initiator ID only when you are in cluster mode.  
You cannot change the ID while in standard mode. The ID can be a  
number from 0 to 15. It is recommended that you use 6 or 7. When  
you are in standard mode, the ID is always 7.  
After you change the Initiator ID, the system prompts you to reboot.  
10 Reboot the system.  
After you reboot, the Adapter menu displays the new Initiator ID.  
186  
PERC 3 BIOS Configuration Utility  
S EC T IO N 1 2  
Dell Manager  
Starting Dell Manager  
Dell Manager Menu Options  
Designating Drives as Hot Spares  
Parameters  
Initializing Logical Drives  
Deleting Logical Drives (Random Array Deletion)  
Formatting Physical Drives  
Rebuilding Failed Disk Drives  
Exiting Dell Manager  
Starting Dell Manager  
Dell™ Manager is a character-based, non-GUI utility that changes policies,  
and parameters, and monitors RAID systems. Dell Manager runs under  
Novell® NetWare® 5.x, 6.x, and Red Hat Linux 7.x.  
To start Dell Manager from any directory, make sure the program file is in  
the usr/sbin directory. Type the command shown in Table 12-1.  
Table 12-1. Command Used to Start Dell Manager  
In...  
Type this:  
Red Hat Linux,  
Novell NetWare  
Dellmgr  
The first Dell Manager screen displays the Management Menu.  
188  
Dell Manager  
Management Menu Options  
Table 12-2 displays the options for the Dell Manager Management Menu.  
Table 12-2. Management Menu Options  
Option  
Initialize  
Objects  
Description  
Select this option to initialize one or more logical drives.  
Select this option to individually access controllers, logical  
drives, and physical drives.  
Format  
Rebuild  
Select this option to low-level format physical drives.  
Select this option to rebuild failed physical drives.  
Check  
Consistency  
Select this option to verify redundancy data in logical drives  
using RAID level 1 or 5. PERC 3 automatically corrects any  
differences found in the data.  
Reconstruct  
Select this option to reconstruct an array.  
Using Dell Manager in Red Hat Linux GUI Mode  
On a Red Hat Linux system, for Dell Manager to work correctly in a  
terminal in GUI Mode, you must:  
•
•
Set the terminal type to linux.  
Set the keyboard mappings to linux.  
Perform the procedure in this section if you use konsole, gnome terminal, or  
xterm.  
The linux console mode, which you select from the terminal with the File—  
> Linux Console command, works correctly by default. The text mode  
console (non-GUI) also works correctly by default.  
Perform the following steps to prepare the system to use Dell Manager in  
Red Hat Linux GUI mode:  
1
2
Start the Terminal.  
Before you enter dellmgr to start Dell Manager, type the following  
commands:  
TERM=linux  
Export TERM  
Dell Manager  
189  
3
Select Settings—> Keyboard—> Linux Console from the Terminal  
menu.  
Dell Manager Menu Options  
The following sections list the menu and sub-menu options available from  
the Dell Manager Management Menu.  
Initialize Menu  
Choose Initialize from the Dell Manager Management Menu to initialize  
one or more logical drives. This action typically follows the configuration of  
a new logical drive.  
Objects Menu  
Choose Objects from the Dell Manager Management Menu to access the  
controllers, logical drives, physical drives, and SCSI channels individually.  
You can also change certain settings for each object. The Objects menu  
options are described in Table 12-3.  
Adapter  
Select Objects—> Adapter select a PERC controller (if your computer has  
more than one) and modify parameters. You can install only one PERC 3  
controller, but you can install other PERC controllers in the computer.  
Table 12-3 displays the Adapter menu options.  
Table 12-3. Adapter Menu Options  
Option  
Description  
FlexRAID  
PowerFail  
Select this option to allow drive reconstruction to continue  
when the system restarts if a power failure occurs. This will  
reduce the size of the logical drive by [1 MB times the  
number of drives in the logical drive].  
Disk Spin up  
Timings  
Select this option to set the method and timing for spinning  
up the hard drives in the computer.  
Alarm Control  
Select this option to enable, disable, or silence the onboard  
alarm tone generator.  
190  
Dell Manager  
Table 12-3. Adapter Menu Options (continued)  
Option  
Description  
View/Update  
Parameters  
Select this option to display the firmware version number and  
the cache memory size. You can also change the rebuild rate  
for the adapter through this option.  
Adapter Type  
The model of adapter that is being used.  
Logical Drive  
Choose Objects—> Logical Drive to select a logical drive and to perform  
the actions described in Table 12-4.  
Table 12-4. Logical Drive Menu Options  
Option  
Description  
Initialize  
Select this option to initialize the selected logical drive. This  
should be done for every logical drive you configure.  
Check  
Consistency  
Select this option to verify the correctness of the redundancy  
data in the selected logical drive. This option is available only if  
RAID level 1 or 5 is used. PERC 3 automatically corrects any  
differences found in the data.  
View/Update  
Parameters  
Select this option to display the properties of the selected logical  
drive. You can modify the cache write policy, the cache read  
policy, virtual sizing and the I/O policy from this menu.  
Virtual Sizing: Set this option to Enabled before adding a  
physical drive to a logical drive. After you have created a logical  
drive set, the partition of the drive should be as large as the  
virtual size of the logical drive.  
Logical Drive Settings  
Table 12-5 displays the default logical drive settings. You can change the  
write policy, read policy, and cache policy after configuration is complete.  
Table 12-5. Logical Drive Default Settings  
Parameter  
Stripe Size  
Write Policy  
Setting  
64 KB  
Write-back  
Dell Manager  
191  
Table 12-5. Logical Drive Default Settings  
Parameter  
Read Policy  
Cache Policy  
Spanning  
Setting  
Adaptive Readahead  
Direct I/O  
Off  
Physical Drive  
Select Objects—> Physical Drive to select a physical device and to perform  
the operations listed in Table 12-5. When you choose this option, the  
physical drives in the system are listed. Move the cursor to the desired  
device and press <Enter> to display the options for physical drives.  
Physical Drive Menu  
Table 12-6 displays the options in the physical drive menu.  
Table 12-6. Physical Drive Menu Options  
Option  
Rebuild  
Format  
Description  
Choose this option to rebuild the selected physical drive.  
Choose this option to low-level format the selected physical  
drive.  
Force Online  
Force Offline  
Choose this option to change the state of the selected  
physical drive to online.  
Choose this option to change the state of the selected  
physical drive to offline.  
Prepare to  
Remove  
Choose this option before you remove a physical drive from  
an array.  
Make HotSpare  
Choose this option to designate the selected physical drive as  
a hot spare.  
View Drive  
Information  
Choose this option to see the manufacturer data for the  
selected physical device.  
View Rebuild  
Progress  
Choose this option to see the progress of the rebuild process  
for the selected physical drive.  
192  
Dell Manager  
SCSI Channel  
Choose Objects—> SCSI Channel to select a SCSI channel on a controller.  
Table 12-7 displays the operations you can perform on the selected channel.  
Table 12-7. SCSI Channel Options  
Option  
Description  
State  
Choose this option to enable termination.  
Choose this option to set the type of termination.  
Termination  
Battery Backup  
Choose Objects-> Battery Backup to view the battery backup information.  
You can use the battery backup menu to reset the charge cycle count. Table  
12-8 displays the menu items.  
Table 12-8. Batter y Backup Menu Items  
Menu Item  
Explanation  
Backup  
Module  
PRESENT will appear if there is a battery module; ABSENT if  
there is not.  
Battery Pack  
Temperature  
Voltage  
PRESENT will appear if the battery pack is properly installed;  
ABSENT if it is not.  
GOOD appears if the temperature is within the normal range.  
HIGH appears if the module is too hot.  
GOOD appears if the voltage is within the normal range. BAD  
appears if the voltage is out of range.  
Fast Charging  
COMPLETED appears if the fast charge cycle is done.  
CHARGING appears if the battery pack is charging.  
Dell Manager  
193  
Table 12-8. Batter y Backup Menu Items (continued)  
Menu Item  
Explanation  
No of Cycles  
This must be configured. When first installing a battery pack,  
set the Charge Cycle to 0. The dialog box below appears when  
you select No of Cycles. Choose YES to reset the number of  
cycles to zero and press <Enter>.  
After 1100 charge cycles, the life of the battery pack is assumed  
to be over, and you must replace it.  
Format Menu  
Choose Format from the Dell Manager Management Menu to low-level  
format one or more physical drives.  
Because most SCSI disk drives are low-level formatted at the factory, this  
step is usually not necessary. You typically must format a disk if:  
•
•
The disk drive was not low-level formatted at the factory, or  
An excessive number of media errors has been detected on the disk  
drive.  
You do not need to use the Format option if you simply want to erase  
existing information on your SCSI disks, such as a DOS partition. That  
information is erased when you initialize the logical drive(s).  
194  
Dell Manager  
Rebuild Menu  
Choose Rebuild from the Management Menu to rebuild one or more failed  
disk drives.  
Check Consistency Menu  
Choose this option to verify the redundancy data in logical drives using  
RAID levels 1 or 5. When you choose Check Consistency, the parameters of  
the existing logical drives on the current controller appear. The logical drives  
are listed by number. PERC 3 automatically corrects any differences found  
in the data. Perform the following steps to run a consistency check.  
1
2
Press the arrow keys to select the desired logical drives.  
Press the spacebar to select or deselect a drive for the consistency  
check.  
3
4
Press <F2> to select or deselect all the logical drives.  
Press <F10> to begin the consistency check.  
A progress indicator for each selected logical drive appears.  
5
6
When the consistency check is finished, press any key to clear the  
progress display.  
Press <Esc> to display the Management Menu.  
Reconstruct Menu  
You can select or deselect a logical drive to be recreated using this option.  
Choose this option before you add or remove a physical drive.  
Designating Drives as Hot Spares  
Hot spares are physical drives that are powered up along with the RAID  
drives and usually stay in a standby state. If a physical drive used in a RAID  
logical drive fails, a hot spare will automatically take its place and the data  
on the failed drive is reconstructed on the hot spare. Hot spares can be used  
for RAID levels 1 and 5. Each PERC 3 controller supports up to eight hot  
spares. See Chapter 2 "Introduction to RAID" for an explanation of hot  
spares.  
Perform the following steps to designate physical drives as hot spares:  
Dell Manager  
195  
1
2
Select Objects—> Physical Drive and press <Enter>.  
The Physical Drive Selection Menu displays.  
Highlight a drive on the Physical Drive Selection Menu using the  
arrow keys and press <Enter>.  
The action menu for the physical drive displays.  
3
Use the arrow key to select Make HotSpare, then press <Enter>.  
The indicator for the selected physical drive changes to HOTSP.  
Parameters  
The parameters contain the settings for the stripe size, write policy, read  
policy and cache policy. The policies and default settings are described in  
the following information.  
•
Stripe Size specifies the size of the segment written to each disk in a  
RAID 1 or 5 logical drive. You can set the stripe size to 2 KB, 4 KB, 8  
KB, 16 KB, 32 KB, 64 KB, or 128 KB. A larger stripe size produces  
better read performance, especially if your computer does mostly  
sequential reads. If you are sure that your computer does random reads  
more often, select a small stripe size. The default stripe size is 64 KB.  
•
Write Policy specifies the cache write policy. You can set the write  
policy to write-back or write-through.  
–
In Write-back caching, the controller sends a data transfer  
completion signal to the host when the controller cache has  
received all the data in a transaction. This is the default setting.  
–
In Write-through caching, the controller sends a data transfer  
completion signal to the host when the disk subsystem has  
received all the data in a transaction.  
Write-through caching has a data security advantage over write-back  
caching. Write-back caching has a performance advantage over write-  
through caching.  
•
Read Policy specifies that the SCSI read-ahead feature is used for the  
logical drive. You can set this parameter to Normal, Read-ahead, or  
Adaptive. The default setting is Adaptive.  
–
Normal specifies that the controller does not use read-ahead for  
the current logical drive.  
196  
Dell Manager  
–
–
Read-ahead specifies that the controller uses read-ahead for the  
current logical drive.  
Adaptive specifies that the controller begins using read-ahead if  
the two most recent disk accesses occurred in sequential sectors. If  
all read requests are random, the algorithm reverts to Normal,  
however, all requests are still evaluated for possible sequential  
operation.  
•
Cache Policy applies to reads on a specific logical drive. It does not  
affect the Read-ahead cache.  
–
–
Cached I/O specifies that all reads are buffered in cache memory.  
Direct I/O specifies that reads are not buffered in cache memory.  
This is the default setting. Direct I/O does not override the cache  
policy settings. Data is transferred to cache and the host  
concurrently. If the same data block is read again, it comes from  
cache memory.  
Initializing Logical Drives  
You should initialize each new logical drive you configure. You can initialize  
the logical drives in two ways:  
•
Batch Initialization. The Initialize option on the Management Menu  
lets you initialize up to eight logical drives simultaneously.  
•
Individual Initialization. The Objects—> Logical Drive action menu  
for an individual logical drive has an Initialize option.  
Batch Initialization  
Perform the following steps to initialize more than one logical drive.  
1
Select Initialize from the Dell Manager Management Menu.  
A list of the current logical drives appears.  
2
3
Press the arrow keys to highlight the drives to be initialized.  
Press the spacebar to select the desired logical drives for initialization  
or press <F2> to select or deselect all the logical drives.  
4
5
After you select the logical drives, press <F10>.  
Select Yes at the confirmation prompt.  
Dell Manager  
197  
The progress of the initialization for each drive displays in bar graph  
format.  
6
7
When initialization is complete, press any key to continue.  
Press <Esc> to display the Management Menu.  
Individual Initialization  
Perform the following steps to initialize one logical drive.  
1
2
3
Select Objects—> Logical Drive.  
Select the logical drive to be initialized.  
Choose Initialize from the action menu.  
The progress of the initialization displays as a graph on the screen.  
4
When initialization completes, press any key to display the previous  
menu.  
Deleting Logical Drives (Random Array  
Deletion)  
This RAID controller supports random array deletion. Random array  
deletion is the ability to delete any unwanted logical drives and use that  
space for a new logical drive.  
NOTE: For more  
information about  
random array deletion,  
refer to page 99.  
The main benefit of random array deletion on the configuration module is  
that you are not restricted to sequential or contiguous logical drives when  
you create logical drives. You can use non-contiguous segments to create  
logical drives.  
After you delete a logical drive, you can create a new one. You can use the  
configuration utilities to create the next logical drive from the non-  
contiguous free space (‘holes’), and from the newly created arrays. The  
configuration utility provides a list of configurable arrays where there is a  
space to configure.  
NOTICE: The deletion of the logical drive can fail under certain conditions.  
You cannot delete a logical drive during a reconstruction. Deletion can fail  
during a rebuild, initialization or check consistency of a logical drive, if that  
drive has a higher logical drive number than the drive you want to delete.  
198  
Dell Manager  
Perform the following steps to delete logical drives:  
1
Choose Objects—> Logical Drive.  
The logical drives display.  
2
3
Use the arrow key to highlight the logical drive you want to delete.  
Press <F5> to delete the logical drive.  
This deletes the logical drive and makes the space it occupied available  
for you to make another logical drive.  
Formatting Physical Drives  
You can do low-level formatting of SCSI drives using Dell Manager. Because  
most SCSI disk drives are low-level formatted at the factory, this step is  
usually not necessary. You typically must format a disk if:  
•
•
The disk drive was not low-level formatted at the factory, or  
An excessive number of media errors have been detected on the disk  
drive.  
Media Errors  
Perform the following steps to check the properties screen for media errors  
for the drive you wish to format.  
1
2
3
Select Physical Drive—> Objects.  
Press the arrow keys to select the desired drive.  
Press <F2>.  
The error count appears at the bottom of the properties screen. If you  
feel that the number of errors is excessive, you should probably format  
the disk drive. If more than 32 media errors were detected, PERC 3  
automatically puts the drive in FAIL state. In that case, formatting the  
drive can clear up the problem.  
You do not have to use the Format option to erase existing information  
on your SCSI disks, such as a DOS partition. That information is  
erased when you initialize logical drives.  
Dell Manager  
199  
Formatting Drives  
You can format the physical drives using:  
•
•
Batch Formatting The Format option in the Dell Manager  
Management Menu lets you format up to eight disk drives  
simultaneously.  
Individual Formatting. Choose the Format option from the  
Objects—> Physical Drive options menu for an individual disk drive.  
Batch Formatting  
1
Choose Format on the Management Menu.  
A device selection window displays the devices connected to the  
current controller.  
2
Press the arrow keys to select the drives to be formatted.  
The indicators for selected drives flash.  
3
4
When you have selected the physical drives, press <F10>.  
Select Yes to the confirmation prompt to begin formatting.  
The indicators for the selected drives change to FRMT[number],  
where number reflects the drive selection order. Formatting can take  
some time, depending on the number of drives you have selected and  
the drive capacities.  
5
6
When formatting is complete, press any key to continue.  
Press <Esc> to display the Management Menu.  
Individual Formatting  
1
2
3
Select Objects—> Physical Drive on the Management Menu.  
A device selection window displays the devices connected to the  
current controller.  
Press the arrow keys to select the physical drive to be formatted and  
press <Enter>.  
The physical drive action menu displays.  
Choose the Format option from the physical drive action menu and  
respond to the confirmation prompt.  
Formatting can take some time, depending on the drive capacity.  
200  
Dell Manager  
4
When formatting completes, press any key to display the previous  
menu.  
Rebuilding Failed Disk Drives  
If a physical drive fails in an array configured as a RAID 1 or 5 logical drive,  
you can recover the lost data by rebuilding the drive. If a rebuilding spare  
fails, a new rebuild is started using a second spare, if available. The capacity  
of the second spare must be equal to or greater than the failed drive.  
Table 12-9 describes the rebuild types.  
Table 12-9. Rebuild Types  
Type  
Description  
Automatic If you have configured hot spares, PERC 3 automatically tries to  
Rebuild  
use them to rebuild failed disks. Display the Objects—> Physical  
Drive screen while a rebuild is in progress. The drive indicator for  
the hot spare disk drive has changed to REBLD A[array number]-  
[drive number], indicating the disk drive being replaced by the hot  
spare.  
Manual  
Rebuild  
Manual rebuild is necessary if no hot spares with enough capacity to  
rebuild the failed drives are available. Select the Rebuild option on  
the Management Menu or the Objects—> Physical Drive menu.  
Manual Rebuild – Rebuilding an Individual Drive  
Perform the following steps to rebuild one drive:  
1
Choose Objects—> Physical Drive on the Management Menu.  
A window displays the devices connected to the current controller.  
2
3
Press the arrow keys to select the physical drive to be rebuilt and press  
<Enter>.  
Choose Rebuild from the physical drive action menu and respond to  
the confirmation prompt.  
Rebuilding can take some time, depending on the drive capacity.  
When rebuild completes, press any key to display the previous menu.  
4
Dell Manager  
201  
Manual Rebuild – Batch Mode  
Perform the following steps to rebuild more than one drive:  
1
Choose Rebuild from the Management Menu.  
A device selection window displays the devices connected to the  
current controller. The failed drives have FAIL indicators.  
2
3
4
Press the arrow keys to select the drives to be rebuilt.  
After you select the physical drives, press <F10>.  
Type Yes at the confirmation prompt.  
The indicators for the selected drives change to REBLD. Rebuilding  
can take some time, depending on the number of drives you have  
selected and their capacities.  
5
6
When rebuild is complete, press any key to continue.  
Press <Esc> to display the Management Menu.  
Exiting Dell Manager  
To exit Dell Manager, press <Esc> from the main menu and choose Yes at  
the prompt. A message appears if uninitialized logical drives remain in the  
system.  
202  
Dell Manager  
S EC T IO N 1 3  
Troubleshooting  
Overview  
BIOS Boot Error Messages  
Other BIOS Error Messages  
Other Potential Problems  
Audible Warnings  
Overview  
This chapter describes the problems that you might encounter during  
installation and use of PERC, and how to resolve them. Table 13-1 describes  
general problems you might encounter.  
Table 13-1. General Problems and Suggested Solutions  
Problem  
Suggested Solution  
Some operating systems do  
not load in a computer with  
a PERC 3 adapter.  
Check the system basic input/output system  
(BIOS) configuration for PCI interrupt  
assignments. Make sure a unique interrupt is  
assigned for PERC 3. Initialize the logical drive  
before installing the operating system.  
One of the hard drive in the  
array fails often.  
Check the drive error counts.  
Format the drive.  
Rebuild the drive.  
If the drive continues to fail, replace the drive  
with another drive with the same capacity.  
Also, check the SCSI cables.  
Pressed <Ctrl> <M> and  
tried to make a new  
configuration. The system  
hangs when scanning  
devices.  
Check the drives IDs on each channel to make  
sure each device has a different ID.  
Check the termination. The device at the end of  
the channel must be terminated.  
Replace the drive cable.  
Multiple drives connected to  
PERC 3 using the same  
power supply. There is a  
problem spinning the drives  
all at once.  
Set the drives to spin on command. This will  
allow PERC 3 to spin two devices  
simultaneously.  
Pressing <Ctrl> <M> does  
not display a menu.  
These utilities require a color monitor.  
At system power-up with the  
PERC 3 installed, the PERC  
3 BIOS banner display is  
garbled or does not appear at  
all.  
PERC 3 cache memory may be defective or  
missing.  
Cannot flash or update the  
EEPROM.  
Contact Dell™ support.  
204  
Troubleshooting  
Table 13-1. General Problems and Suggested Solutions  
Problem  
Suggested Solution  
The following displays:  
Make sure that termination power (TERMPWR)  
is being properly provided to each peripheral  
device- populated channel.  
Firmware  
Initializing...  
Make sure that each end of the SCSI channel  
chain is properly terminated using the  
recommended terminator type for the peripheral  
device. The channel is automatically terminated  
at the PERC 3 card if only one cable is  
connected to a channel.  
Make sure that the PERC 3 controller is properly  
seated in the PCI slot.  
Why does a failed logical  
array still get a drive  
assignment?  
To maintain the integrity of the DOS path  
statement.  
BIOS Boot Error Messages  
Table 13-2 explains the error messages that can display when there is an  
error during bootup.  
Table 13-2. BIOS Boot Error Messages  
Message  
Problem  
Suggested Solution  
Adapter BIOS  
Disabled. No  
Logical Drives  
Handled by BIOS  
The PERC 3 BIOS is  
disabled. Sometimes  
the BIOS is disabled  
to prevent booting  
from the BIOS. This is  
the default when  
cluster mode is  
Enable the BIOS by pressing  
<Ctrl> <M> at the boot  
prompt to run the BIOS  
Configuration Utility.  
enabled.  
Host Adapter at  
Baseport xxxx  
Not Responding  
The BIOS cannot  
communicate with the  
adapter firmware.  
Make sure PERC 3 is  
properly installed. Check  
SCSI termination and  
cables.  
No PERC 3  
Adapter  
The BIOS cannot  
communicate with the  
adapter firmware.  
Make sure PERC 3 is  
properly installed.  
Troubleshooting  
205  
Table 13-2. BIOS Boot Error Messages (continued)  
Message  
Problem  
Suggested Solution  
Configuration  
of NVRAM and  
drives  
mismatch. Run  
View/Add  
The configuration  
stored in the PERC 3  
adapter does not  
match the  
Press <Ctrl><M> to run  
the BIOS Configuration  
Utility. Choose Configure—  
> View/Add Configuration  
configuration stored in to examine both the  
Configuration  
option of PERC  
3 BIOS  
the drives.  
configuration in non-volatile  
random access memory  
(NVRAM) and the  
Configuration  
Utility.  
configuration stored on the  
hard drives. Resolve the  
problem by selecting one of  
the configurations.  
Press any key  
to run the  
Configuration  
Utility.  
Unresolved  
configuration  
mismatch  
between disks  
and NVRAM on  
the adapter  
after creating  
a new  
Some legacy  
Clear the configuration.  
configurations in the  
drives cannot be  
cleared.  
Low-level format the related  
drives and re-create the  
configuration.  
configuration  
1 Logical Drive  
Failed  
A logical drive failed  
to sign on.  
Make sure all physical drives  
are properly connected and  
are powered on. Run the  
BIOS Configuration Utility  
to find out if any physical  
drives are not responding.  
Reconnect, replace, or  
rebuild any drive that is not  
responding.  
TBBU Cache-data  
seems to be  
INVALID for  
current  
The data on the  
transportable battery  
backup (TBBU) unit is you want to discard the  
not valid for the  
Power off and put the TBBU  
on the proper adapter. Or if  
Cache-data and proceed,  
press Y:.  
adapter.  
current controller.  
206  
Troubleshooting  
Table 13-2. BIOS Boot Error Messages (continued)  
Message  
Problem  
Suggested Solution  
X Logical  
Drives Degraded  
x number of logical  
drives signed on in a  
degraded state.  
Make sure all physical drives  
are properly connected and  
are powered on. Run the  
BIOS Configuration Utility  
to find whether any physical  
drives are not responding.  
Reconnect, replace, or  
rebuild a drive that is not  
responding.  
1 Logical Drive  
Degraded  
A logical drive signed  
on in a degraded state. are properly connected and  
are powered on. Run the  
Make sure all physical drives  
BIOS Configuration  
Manager to find out if any  
physical drives are not  
responding. Reconnect,  
replace, or rebuild any drive  
that is not responding.  
Insufficient  
memory to run  
BIOS. Press any  
key to  
Not enough PERC 3  
memory to run PERC  
3 BIOS.  
Make sure PERC 3 cache  
memory has been properly  
installed.  
continue…  
Insufficient  
Memory  
Not enough memory  
on the PERC 3  
adapter to support the  
current configuration.  
Make sure PERC 3 cache  
memory has been properly  
installed.  
The following  
SCSI IDs are  
not responding:  
The physical drives  
with SCSIO IDs a, b,  
and c are not  
Make sure the physical drives  
are properly connected and  
are powered on.  
responding on SCSI  
channel x.  
Channel x:a.b.c  
Troubleshooting  
207  
Other BIOS Error Messages  
Table 13-3 explains the other BIOS error messages that can display.  
Table 13-3. Other BIOS Error Messages  
Message  
Problem  
Suggested Solution  
Following  
The physical disk roaming  
feature did not find the  
physical disk with the  
displayed SCSI ID. No  
slot is available to map the  
physical drive. PERC 3  
cannot resolve the  
Reconfigure the array.  
SCSI disk not  
found and no  
empty slot  
available for  
mapping it  
physical drives into the  
current configuration.  
Following  
SCSI IDs have  
the same data  
y, z  
The physical drive  
Remove the drive or drives  
that should not be used.  
roaming feature found the  
same data on two or more  
physical drives on channel  
x with SCSI IDs a, b, and  
c. PERC 3 cannot  
determine the drive that  
has the duplicate  
Channel x: a,  
b, c  
information.  
Unresolved  
configuration  
mismatch  
The PERC 3 is unable to  
determine the proper  
configuration after  
Press <Ctrl><M> to run  
the BIOS Configuration  
Utility. Choose  
between disks  
and NVRAM on  
the adapter  
reading both NVRAM and  
Configuration on Disk.  
Configure—> New  
Configuration to create a  
new configuration.  
CAUTION: This will delete  
any existing configuration.  
208  
Troubleshooting  
Other Potential Problems  
Table 13-4 describes other items that might cause problems.  
Table 13-4. Other Potential Problems  
Topic  
Information  
Physical drive  
errors  
To display the Media Error and Other Error options, select  
the Objects—> Physical Drive menu and press <F2> after you  
select a physical drive. A Media Error is an error that occurred  
while actually transferring data. An Other Error is an error that  
occurs at the hardware level, such as a device failure, poor  
cabling, bad termination, or signal loss.  
Virtual sizing  
The virtual sizing option enables RAID expansion. Virtual  
sizing must be enabled to increase the size of a logical drive or  
add a physical drive to an existing logical drive. To enable  
virtual sizing, run the BIOS Configuration Utility by pressing  
<Ctrl> <M> at bootup. Select Objects—> Logical Drive.  
Then select View—> Update Parameters. Set virtual sizing to  
Enabled.  
PERC 3 Power  
Requirements  
The maximum PERC 3 power requirements are 15 watts at  
5 v and 3 amps.  
®
Windows NT  
does not detect  
the PERC 3  
Refer to the PERC 3 Operating System Driver Installation  
Guide for the chapter about Windows NT driver installation.  
Troubleshooting  
209  
Audible Warnings  
The PERC card onboard tone generator indicates events and errors. Table  
13-5 describes the warnings and their meanings.  
Table 13-5. Audible Warnings  
Tone Pattern  
Meaning  
Examples  
Three seconds on  
and one second off  
A logical drive is  
offline.  
One or more drives in a RAID 0  
configuration failed.  
Two or more drives in a RAID 1 or  
5 configuration failed.  
One second on and  
one second off  
A logical drive is  
running in  
degraded mode.  
One drive in a RAID 5  
configuration failed.  
One second on and  
three seconds off  
An automatically  
initiated rebuild  
has been  
While you were away from the  
system, a disk drive in a RAID 1  
or 5 configuration failed and was  
rebuilt.  
completed.  
210  
Troubleshooting  
S EC T IO N 1 4  
Appendix A: SCSI  
Cables and Connectors  
Visit Dell’s web site at www.dell.com for information about qualified SCSI cables and connectors, or contact your  
Dell representative for information.  
212  
Appendix A: SCSI Cables and Connectors  
S EC T IO N 1 5  
Appendix B:  
Warranties and Return  
Policy  
Dell-branded hardware products purchased in the U.S. or Canada come  
with either a one-year, two-year, three-year, or four-year (U.S. only) limited  
warranty. To determine which warranty you purchased, see the Dell invoice  
that accompanied your computer. The following sections describe these  
limited warranties, the manufacturer guarantee for Latin America and the  
Caribbean, and the return policy for the U.S.  
One-Year Limited Warranty (U.S. Only)  
Dell Computer Corporation ("Dell") manufactures its hardware products  
from parts and components that are new or equivalent to new in accordance  
with industry-standard practices. Dell warrants that the hardware  
products it manufactures will be free from defects in materials and  
workmanship. The limited warranty term is one year beginning on the date  
of invoice, as further described in the following text.  
Damage due to shipping the products to you is covered under this limited  
warranty. Otherwise, this limited warranty does not cover damage due to  
external causes, including accident, abuse, misuse, problems with electrical  
power, servicing not authorized by Dell, usage not in accordance with  
product instructions, failure to perform required preventive maintenance,  
and problems caused by use of parts and components not supplied by Dell.  
This limited warranty does not cover any items that are in one or more of  
the following categories: software; external devices (except as specifically  
noted); accessories or parts added to a Dell system after the system is  
shipped from Dell; accessories or parts added to a Dell system through  
Dell’s system integration department; accessories or parts that are not  
installed in the Dell factory; or Dell Software and Peripherals products.  
Monitors, keyboards, and mice that are Dell-branded or that are included  
on Dell’s standard price list are covered under this limited warranty; all  
other monitors, keyboards, and mice (including those sold through the Dell  
Software and Peripherals program) are not covered.  
214  
Appendix B: Warranties and Return Policy  
Dell will repair or replace products covered under this limited warranty that  
are returned to Dell’s facility. To request warranty service, you must contact  
Dell’s Customer Technical Support within the limited warranty period. If  
warranty service is required, Dell will issue a Return Material Authorization  
Number. You must ship the products back to Dell in their original or  
equivalent packaging, prepay shipping charges, and insure the shipment or  
accept the risk of loss or damage during shipment. Dell will ship the  
repaired or replacement products to you (freight prepaid) if you use an  
address in the continental U.S., where applicable. Shipments to other  
locations will be made freight collect.  
NOTE: Before you ship  
the product(s) to Dell,  
back up the data on the  
hard drive(s) and any  
other storage device(s) in  
the product(s). Remove  
any removable media,  
such as diskettes, CDs, or  
PC Cards. Dell does not  
accept liability for lost  
data or software.  
Dell owns all parts removed from repaired products. Dell uses new and  
reconditioned parts made by various manufacturers in performing warranty  
repairs and building replacement products. If Dell repairs or replaces a  
product, its warranty term is not extended.  
THIS LIMITED WARRANTY GIVES YOU SPECIFIC LEGAL RIGHTS,  
AND YOU MAY ALSO HAVE OTHER RIGHTS, WHICH VARY FROM  
STATE TO STATE (OR JURISDICTION TO JURISDICTION). DELL'S  
RESPONSIBILITY FOR MALFUNCTIONS AND DEFECTS IN  
HARDWARE IS LIMITED TO REPAIR AND REPLACEMENT AS SET  
FORTH IN THIS LIMITED WARRANTY STATEMENT. ALL EXPRESS  
AND IMPLIED WARRANTIES FOR THE PRODUCT, INCLUDING  
BUT NOT LIMITED TO ANY IMPLIED WARRANTIES OF AND  
CONDITIONS OF MERCHANTABILITY AND FITNESS FOR A  
PARTICULAR PURPOSE, ARE LIMITED IN DURATION TO THE  
WARRANTY PERIOD SET FORTH ABOVE AND NO WARRANTIES,  
WHETHER EXPRESS OR IMPLIED, WILL APPLY AFTER SUCH  
PERIOD.  
SOME STATES (OR JURISDICTIONS) DO NOT ALLOW  
LIMITATIONS ON HOW LONG AN IMPLIED WARRANTY LASTS, SO  
THE PRECEDING LIMITATION MAY NOT APPLY TO YOU.  
DELL DOES NOT ACCEPT LIABILITY BEYOND THE REMEDIES  
SET FORTH IN THIS LIMITED WARRANTY STATEMENT OR  
LIABILITY FOR INCIDENTAL OR CONSEQUENTIAL DAMAGES,  
INCLUDING WITHOUT LIMITATION ANY LIABILITY FOR  
PRODUCTS NOT BEING AVAILABLE FOR USE OR FOR LOST DATA  
OR SOFTWARE.  
Appendix B: Warranties and Return Policy  
215  
SOME STATES (OR JURISDICTIONS) DO NOT ALLOW THE  
EXCLUSION OR LIMITATION OF INCIDENTAL OR  
CONSEQUENTIAL DAMAGES, SO THE PRECEDING EXCLUSION  
OR LIMITATION MAY NOT APPLY TO YOU.  
These provisions apply to Dell’s one-year limited warranty only. For  
provisions of any service contract covering your system, see your invoice or  
the separate service contract that you will receive.  
If Dell elects to exchange a product or portion of a product, the exchange  
will be made in accordance with Dell’s Exchange Policy in effect on the date  
of the exchange. In any instance in which Dell issues a Return Material  
Authorization Number, Dell must receive the product(s) for repair prior to  
the expiration of the warranty period in order for the repair(s) to be covered  
by the limited warranty.  
Two-Year Limited Warranty (U.S. Only)  
Dell Computer Corporation ("Dell") manufactures its hardware products  
from parts and components that are new or equivalent to new in accordance  
with industry-standard practices. Dell warrants that the hardware products  
it manufactures will be free from defects in materials and workmanship.  
The limited warranty term is two years beginning on the date of invoice, as  
further described in the following section.  
Damage due to shipping the products to you is covered under this warranty.  
Otherwise, this limited warranty does not cover damage due to external  
causes, including accident, abuse, misuse, problems with electrical power,  
servicing not authorized by Dell, usage not in accordance with product  
instructions, failure to perform required preventive maintenance, and  
problems caused by use of parts and components not supplied by Dell.  
216  
Appendix B: Warranties and Return Policy  
This limited warranty does not cover any items that are in one or more of  
the following categories: software; external devices (except as specifically  
noted); accessories or parts added to a Dell system after the system is  
shipped from Dell; accessories or parts added to a Dell system through  
Dell's system integration department; accessories or parts that are not  
installed in the Dell factory; or Dell Software and Peripherals products.  
Monitors, keyboards, and mice that are Dell-branded or that are included  
on Dell's standard price list are covered under this warranty; all other  
monitors, keyboards, and mice (including those sold through the Dell  
Software and Peripherals program) are not covered. Batteries for portable  
computers are covered only during the initial one-year period of this limited  
warranty.  
Limited Warranty Coverage During Year One  
During the one-year period beginning on the invoice date, Dell will repair or  
replace products covered under this limited warranty that are returned to  
Dell's facility. To request warranty service, you must contact Dell’s  
Customer Technical Support within the warranty period. If warranty service  
is required, Dell will issue a Return Material Authorization Number. You  
must ship the products back to Dell in their original or equivalent  
packaging, prepay shipping charges, and insure the shipment or accept the  
risk of loss or damage during shipment. Dell will ship the repaired or  
replacement products to you freight prepaid if you use an address in the  
continental U.S., where applicable. Shipments to other locations will be  
made freight collect.  
NOTE: Before you ship  
the product(s) to Dell,  
back up the data on the  
hard drive(s) and any  
other storage device(s) in  
the product(s). Remove  
any removable media,  
such as diskettes, CDs, or  
PC Cards. Dell does not  
accept liability for lost  
data or software.  
Dell owns all parts removed from repaired products. Dell uses new and  
reconditioned parts made by various manufacturers in performing warranty  
repairs and building replacement products. If Dell repairs or replaces a  
product, its warranty term is not extended.  
Appendix B: Warranties and Return Policy  
217  
Limited Warranty Coverage During Year Two  
During the second year of this limited warranty, Dell will provide, on an  
exchange basis and subject to Dell’s Exchange Policy in effect on the date of  
the exchange, replacement parts for the Dell hardware product(s) covered  
under this limited warranty when a part requires replacement. You must  
report each instance of hardware failure to Dell’s Customer Technical  
Support in advance to obtain Dell's concurrence that a part should be  
replaced and to have Dell ship the replacement part. Dell will ship parts  
(freight prepaid) if you use an address in the continental U.S., where  
applicable. Shipments to other locations, where applicable, will be made  
freight collect. Dell will include a prepaid shipping container with each  
replacement part for your use in returning the replaced part to Dell.  
Replacement parts are new or reconditioned. Dell may provide replacement  
parts made by various manufacturers when supplying parts to you. The  
warranty term for a replacement part is the remainder of the limited  
warranty term.  
You will pay Dell for replacement parts if the replaced part is not returned to  
Dell. The process for returning replaced parts, and your obligation to pay for  
replacement parts if you do not return the replaced parts to Dell, will be in  
accordance with Dell’s Exchange Policy in effect on the date of the  
exchange.  
You accept full responsibility for your software and data. Dell is not required  
to advise or remind you of appropriate backup and other procedures.  
218  
Appendix B: Warranties and Return Policy  
General Provisions  
THIS LIMITED WARRANTY GIVES YOU SPECIFIC LEGAL RIGHTS,  
AND YOU MAY ALSO HAVE OTHER RIGHTS, WHICH VARY FROM  
STATE TO STATE (OR JURISDICTION TO JURISDICTION). DELL'S  
RESPONSIBILITY FOR MALFUNCTIONS AND DEFECTS IN  
HARDWARE IS LIMITED TO REPAIR AND REPLACEMENT AS SET  
FORTH IN THIS LIMITED WARRANTY STATEMENT. ALL EXPRESS  
AND IMPLIED WARRANTIES FOR THE PRODUCT, INCLUDING  
BUT NOT LIMITED TO ANY IMPLIED WARRANTIES OF AND  
CONDITIONS OF MERCHANTABILITY AND FITNESS FOR A  
PARTICULAR PURPOSE, ARE LIMITED IN DURATION TO THE  
WARRANTY PERIOD SET FORTH ABOVE AND NO WARRANTIES,  
WHETHER EXPRESS OR IMPLIED, WILL APPLY AFTER SUCH  
PERIOD. SOME STATES (OR JURISDICTIONS) DO NOT ALLOW  
LIMITATIONS ON HOW LONG AN IMPLIED WARRANTY LASTS, SO  
THE PRECEDING LIMITATION MAY NOT APPLY TO YOU.  
DELL DOES NOT ACCEPT LIABILITY BEYOND THE REMEDIES  
SET FORTH IN THIS LIMITED WARRANTY STATEMENT OR  
LIABILITY FOR INCIDENTAL OR CONSEQUENTIAL DAMAGES,  
INCLUDING WITHOUT LIMITATION ANY LIABILITY FOR  
PRODUCTS NOT BEING AVAILABLE FOR USE OR FOR LOST DATA  
OR SOFTWARE. SOME STATES (OR JURISDICTIONS) DO NOT  
ALLOW THE EXCLUSION OR LIMITATION OF INCIDENTAL OR  
CONSEQUENTIAL DAMAGES, SO THE PRECEDING EXCLUSION  
OR LIMITATION MAY NOT APPLY TO YOU.  
These provisions apply to Dell's two-year limited warranty only. For  
provisions of any service contract covering your system, see your invoice or  
the separate service contract that you will receive.  
If Dell elects to exchange a system or component, the exchange will be  
made in accordance with Dell’s Exchange Policy in effect on the date of the  
exchange. In any instance in which Dell issues a Return Material  
Authorization Number, Dell must receive the product(s) for repair prior to  
the expiration of the warranty period in order for the repair(s) to be covered  
by the limited warranty.  
Appendix B: Warranties and Return Policy  
219  
Three-Year Limited Warranty (U.S. Only)  
Dell Computer Corporation ("Dell") manufactures its hardware products  
from parts and components that are new or equivalent to new in accordance  
with industry-standard practices. Dell warrants that the hardware products  
it manufactures will be free from defects in materials and workmanship.  
The limited warranty term is three years beginning on the date of invoice, as  
described in the following text.  
Damage due to shipping the products to you is covered under this limited  
warranty. Otherwise, this limited warranty does not cover damage due to  
external causes, including accident, abuse, misuse, problems with electrical  
power, servicing not authorized by Dell, usage not in accordance with  
product instructions, failure to perform required preventive maintenance,  
and problems caused by use of parts and components not supplied by Dell.  
This limited warranty does not cover any items that are in one or more of  
the following categories: software; external devices (except as specifically  
noted); accessories or parts added to a Dell system after the system is  
shipped from Dell; accessories or parts added to a Dell system through  
Dell’s system integration department; accessories or parts that are not  
installed in the Dell factory; or Dell Software and Peripherals products.  
Monitors, keyboards, and mice that are Dell-branded or that are included  
on Dell’s standard price list are covered under this limited warranty; all  
other monitors, keyboards, and mice (including those sold through the Dell  
Software and Peripherals program) are not covered. Batteries for portable  
computers are covered only during the initial one-year period of this limited  
warranty.  
220  
Appendix B: Warranties and Return Policy  
Limited Warranty Coverage During Year One  
During the one-year period beginning on the invoice date, Dell will repair or  
replace products covered under this limited warranty that are returned to  
Dell’s facility. To request warranty service, you must contact Dell’s  
Customer Technical Support within the warranty period. If warranty service  
is required, Dell will issue a Return Material Authorization Number. You  
must ship the products back to Dell in their original or equivalent  
packaging, prepay shipping charges, and insure the shipment or accept the  
risk of loss or damage during shipment. Dell will ship the repaired or  
replacement products to you freight prepaid if you use an address in the  
continental U.S., where applicable. Shipments to other locations will be  
made freight collect.  
NOTE: Before you ship  
the product(s) to Dell,  
back up the data on the  
hard drive(s) and any  
other storage device(s) in  
the product(s). Remove  
any removable media,  
such as diskettes, CDs, or  
PC Cards. Dell does not  
accept liability for lost  
data or software.  
Dell owns all parts removed from repaired products. Dell uses new and  
reconditioned parts made by various manufacturers in performing warranty  
repairs and building replacement products. If Dell repairs or replaces a  
product, its warranty term is not extended.  
Limited Warranty Coverage During Years Two and Three  
During the second and third years of this limited warranty, Dell will provide,  
on an exchange basis and subject to Dell’s Exchange Policy in effect on the  
date of the exchange, replacement parts for the Dell hardware product(s)  
covered under this limited warranty when a part requires replacement. You  
must report each instance of hardware failure to Dell’s Customer Technical  
Support in advance to obtain Dell’s concurrence that a part should be  
replaced and to have Dell ship the replacement part. Dell will ship parts  
(freight prepaid) if you use an address in the continental U.S., where  
applicable. Shipments to other locations will be made freight collect. Dell  
will include a prepaid shipping container with each replacement part for  
your use in returning the replaced part to Dell. Replacement parts are new  
or reconditioned. Dell may provide replacement parts made by various  
manufacturers when supplying parts to you. The warranty term for a  
replacement part is the remainder of the limited warranty term.  
Appendix B: Warranties and Return Policy  
221  
You will pay Dell for replacement parts if the replaced part is not returned to  
Dell. The process for returning replaced parts, and your obligation to pay for  
replacement parts if you do not return the replaced parts to Dell, will be in  
accordance with Dell’s Exchange Policy in effect on the date of the  
exchange.  
You accept full responsibility for your software and data. Dell is not required  
to advise or remind you of appropriate backup and other procedures.  
General Provisions  
THIS LIMITED WARRANTY GIVES YOU SPECIFIC LEGAL RIGHTS,  
AND YOU MAY ALSO HAVE OTHER RIGHTS, WHICH VARY FROM  
STATE TO STATE (OR JURISDICTION TO JURISDICTION). DELL'S  
RESPONSIBILITY FOR MALFUNCTIONS AND DEFECTS IN  
HARDWARE IS LIMITED TO REPAIR AND REPLACEMENT AS SET  
FORTH IN THIS LIMITED WARRANTY STATEMENT. ALL EXPRESS  
AND IMPLIED WARRANTIES FOR THE PRODUCT, INCLUDING  
BUT NOT LIMITED TO ANY IMPLIED WARRANTIES OF AND  
CONDITIONS OF MERCHANTABILITY AND FITNESS FOR A  
PARTICULAR PURPOSE, ARE LIMITED IN DURATION TO THE  
WARRANTY PERIOD SET FORTH ABOVE AND NO WARRANTIES,  
WHETHER EXPRESS OR IMPLIED, WILL APPLY AFTER SUCH  
PERIOD.  
SOME STATES (OR JURISDICTIONS) DO NOT ALLOW  
LIMITATIONS ON HOW LONG AN IMPLIED WARRANTY LASTS, SO  
THE PRECEDING LIMITATION MAY NOT APPLY TO YOU.  
DELL DOES NOT ACCEPT LIABILITY BEYOND THE REMEDIES  
SET FORTH IN THIS LIMITED WARRANTY STATEMENT OR  
LIABILITY FOR INCIDENTAL OR CONSEQUENTIAL DAMAGES,  
INCLUDING WITHOUT LIMITATION ANY LIABILITY FOR  
PRODUCTS NOT BEING AVAILABLE FOR USE OR FOR LOST DATA  
OR SOFTWARE.  
SOME STATES (OR JURISDICTIONS) DO NOT ALLOW THE  
EXCLUSION OR LIMITATION OF INCIDENTAL OR  
CONSEQUENTIAL DAMAGES, SO THE PRECEDING EXCLUSION  
OR LIMITATION MAY NOT APPLY TO YOU.  
222  
Appendix B: Warranties and Return Policy  
These provisions apply to Dell’s three-year limited warranty only. For  
provisions of any service contract covering your system, see your invoice or  
the separate service contract that you will receive.  
If Dell elects to exchange a system or component, the exchange will be  
made in accordance with Dell’s Exchange Policy in effect on the date of the  
exchange. In any instance in which Dell issues a Return Material  
Authorization Number, Dell must receive the product(s) for repair prior to  
the expiration of the warranty period in order for the repair(s) to be covered  
by the limited warranty.  
Four-Year Limited Warranty (U.S. Only)  
Dell Computer Corporation ("Dell") manufactures its hardware products  
from parts and components that are new or equivalent to new in accordance  
with industry-standard practices. Dell warrants that the hardware products  
it manufactures will be free from defects in materials and workmanship.  
The limited warranty term is four years beginning on the date of invoice, as  
further described in the following text.  
Damage due to shipping the products to you is covered under this limited  
warranty. Otherwise, this limited warranty does not cover damage due to  
external causes, including accident, abuse, misuse, problems with electrical  
power, servicing not authorized by Dell, usage not in accordance with  
product instructions, failure to perform required preventive maintenance,  
and problems caused by use of parts and components not supplied by Dell.  
This limited warranty does not cover any items that are in one or more of  
the following categories: software; external devices (except as specifically  
noted); accessories or parts added to a Dell system after the system is  
shipped from Dell; accessories or parts added to a Dell system through  
Dell's system integration department; accessories or parts that are not  
installed in the Dell factory; or Dell Software and Peripherals products.  
Monitors, keyboards, and mice that are Dell-branded or that are included  
on Dell's standard price list are covered under this limited warranty; all other  
monitors, keyboards, and mice (including those sold through the Dell  
Software and Peripherals program) are not covered. Batteries for portable  
computers are covered only during the initial one-year period of this limited  
warranty.  
Appendix B: Warranties and Return Policy  
223  
Limited Warranty Coverage During Year One  
NOTE: Before you ship  
the product(s) to Dell,  
back up the data on the  
hard drive(s) and any  
other storage device(s) in  
the product(s). Remove  
any removable media,  
such as diskettes, CDs, or  
PC Cards. Dell does not  
accept liability for lost  
data or software.  
During the one-year period beginning on the invoice date, Dell will repair or  
replace products covered under this limited warranty that are returned to  
Dell's facility. To request limited warranty service, you must contact Dell's  
Customer Technical Support within the limited warranty period. See  
"Contacting Dell" in your user documentation to find the appropriate  
contact information for obtaining customer assistance. If limited warranty  
service is required, Dell will issue a Return Material Authorization Number.  
You must ship the products back to Dell in their original or equivalent  
packaging, prepay shipping charges, and insure the shipment or accept the  
risk of loss or damage during shipment. Dell will ship the repaired or  
replacement products to you freight prepaid if you use an address in the  
continental U.S., where applicable. Shipments to other locations will be  
made freight collect.  
Dell owns all parts removed from repaired products. Dell uses new and  
reconditioned parts made by various manufacturers in performing limited  
warranty repairs and building replacement products. If Dell repairs or  
replaces a product, its limited warranty term is not extended.  
Limited Warranty Coverage During Years Two, Three, and Four  
During the second, third, and fourth years of this limited warranty, Dell will  
provide, on an exchange basis and subject to Dell's Exchange Policy in effect  
on the date of the exchange, replacement parts for the Dell hardware  
product(s) covered under this limited warranty when a part requires  
replacement. You must report each instance of hardware failure to Dell's  
Customer Technical Support in advance to obtain Dell's concurrence that a  
part should be replaced and to have Dell ship the replacement part. Dell will  
ship parts (freight prepaid) if you use an address in the continental U.S.,  
where applicable. Shipments to other locations will be made freight collect.  
Dell will include a prepaid shipping container with each replacement part  
for your use in returning the replaced part to Dell. Replacement parts are  
new or reconditioned. Dell may provide replacement parts made by various  
manufacturers when supplying parts to you. The limited warranty term for a  
replacement part is the remainder of the limited warranty term.  
224  
Appendix B: Warranties and Return Policy  
You will pay Dell for replacement parts when the replaced part is not  
returned to Dell. The process for returning replaced parts, and your  
obligation to pay for replacement parts if you do not return the replaced  
parts to Dell, will be in accordance with Dell's Exchange Policy in effect on  
the date of the exchange.  
You accept full responsibility for your software and data. Dell is not required  
to advise or remind you of appropriate backup and other procedures.  
General Provisions  
THIS LIMITED WARRANTY GIVES YOU SPECIFIC LEGAL RIGHTS,  
AND YOU MAY ALSO HAVE OTHER RIGHTS, WHICH VARY FROM  
STATE TO STATE (OR JURISDICTION TO JURISDICTION). DELL'S  
RESPONSIBILITY FOR MALFUNCTIONS AND DEFECTS IN  
HARDWARE IS LIMITED TO REPAIR AND REPLACEMENT AS SET  
FORTH IN THIS LIMITED WARRANTY STATEMENT. ALL EXPRESS  
AND IMPLIED WARRANTIES FOR THE PRODUCT, INCLUDING  
BUT NOT LIMITED TO ANY IMPLIED WARRANTIES OF AND  
CONDITIONS OF MERCHANTABILITY AND FITNESS FOR A  
PARTICULAR PURPOSE, ARE LIMITED IN DURATION TO THE  
LIMITED WARRANTY PERIOD SET FORTH ABOVE AND NO  
WARRANTIES, WHETHER EXPRESS OR IMPLIED, WILL APPLY  
AFTER SUCH PERIOD. SOME STATES (OR JURISDICTIONS) DO  
NOT ALLOW LIMITATIONS ON HOW LONG AN IMPLIED  
WARRANTY LASTS, SO THE PRECEDING LIMITATION MAY NOT  
APPLY TO YOU.  
DELL DOES NOT ACCEPT LIABILITY BEYOND THE REMEDIES  
SET FORTH IN THIS LIMITED WARRANTY STATEMENT OR  
LIABILITY FOR INCIDENTAL OR CONSEQUENTIAL DAMAGES,  
INCLUDING WITHOUT LIMITATION ANY LIABILITY FOR  
PRODUCTS NOT BEING AVAILABLE FOR USE OR FOR LOST DATA  
OR SOFTWARE. SOME STATES (OR JURISDICTIONS) DO NOT  
ALLOW THE EXCLUSION OR LIMITATION OF INCIDENTAL OR  
CONSEQUENTIAL DAMAGES, SO THE PRECEDING EXCLUSION  
OR LIMITATION MAY NOT APPLY TO YOU.  
Appendix B: Warranties and Return Policy  
225  
If Dell elects to exchange a system or component, the exchange will be  
made in accordance with Dell's Exchange Policy in effect on the date of the  
exchange. In any instance in which Dell issues a Return Material  
Authorization Number, Dell must receive the product(s) for repair prior to  
the expiration of the limited warranty period in order for the repairs to be  
covered by the limited warranty.  
These provisions apply to Dell's four-year limited warranty only. For  
provisions of any on-site service contract covering your system, see the  
separate on-site service contract that you will receive.  
One-Year Limited Warranty (Canada Only)  
Dell Computer Corporation ("Dell") manufactures its hardware products  
from parts and components that are new or equivalent to new in accordance  
with industry-standard practices. Dell warrants that the hardware  
products it manufactures will be free from defects in materials and  
workmanship. The limited warranty term is one year beginning on the date  
of invoice, as further described in the following text.  
Damage due to shipping the products to you is covered under this limited  
warranty. Otherwise, this limited warranty does not cover damage due to  
external causes, including accident, abuse, misuse, problems with electrical  
power, servicing not authorized by Dell, usage not in accordance with  
product instructions, failure to perform required preventive maintenance,  
and problems caused by use of parts and components not supplied by Dell.  
This limited warranty does not cover any items that are in one or more of  
the following categories: software; external devices (except as specifically  
noted); accessories or parts added to a Dell system after the system is  
shipped from Dell; accessories or parts added to a Dell system through  
Dell’s system integration department; accessories or parts that are not  
installed in the Dell factory; or Dell Software and Peripherals products.  
Monitors, keyboards, and mice that are Dell-branded or that are included  
on Dell’s standard price list are covered under this limited warranty; all  
other monitors, keyboards, and mice (including those sold through the Dell  
Software and Peripherals program) are not covered.  
226  
Appendix B: Warranties and Return Policy  
Dell will repair or replace products covered under this limited warranty that  
are returned to Dell’s facility. To request warranty service, you must contact  
Dell’s Customer Technical Support within the warranty period. If warranty  
service is required, Dell will issue a Return Material Authorization Number.  
You must ship the products back to Dell in their original or equivalent  
packaging, prepay shipping charges, and insure the shipment or accept the  
risk of loss or damage during shipment. Dell will ship the repaired or  
replacement products to you (freight prepaid) if you use an address in  
Canada, where applicable. Shipments to other locations will be made  
freight collect.  
NOTE: Before you ship  
the product(s) to Dell,  
back up the data on the  
hard drive(s) and any  
other storage device(s) in  
the product(s). Remove  
any removable media,  
such as diskettes, CDs, or  
PC Cards. Dell does not  
accept liability for lost  
data or software.  
Dell owns all parts removed from repaired products. Dell uses new and  
reconditioned parts made by various manufacturers in performing warranty  
repairs and building replacement products. If Dell repairs or replaces a  
product, its warranty term is not extended.  
DELL MAKES NO EXPRESS WARRANTIES OR CONDITIONS  
BEYOND THOSE STATED IN THIS LIMITED WARRANTY  
STATEMENT. DELL DISCLAIMS ALL OTHER WARRANTIES AND  
CONDITIONS, EXPRESS OR IMPLIED, INCLUDING WITHOUT  
LIMITATION IMPLIED WARRANTIES AND CONDITIONS OF  
MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.  
SOME STATES (OR JURISDICTIONS) DO NOT ALLOW  
LIMITATIONS ON IMPLIED WARRANTIES OR CONDITIONS, SO  
THIS LIMITATION MAY NOT APPLY TO YOU.  
DELL’S RESPONSIBILITY FOR MALFUNCTIONS AND DEFECTS IN  
HARDWARE IS LIMITED TO REPAIR AND REPLACEMENT AS SET  
FORTH IN THIS LIMITED WARRANTY STATEMENT. THESE  
WARRANTIES GIVE YOU SPECIFIC LEGAL RIGHTS, AND YOU MAY  
ALSO HAVE OTHER RIGHTS, WHICH VARY FROM STATE TO  
STATE (OR JURISDICTION TO JURISDICTION).  
DELL DOES NOT ACCEPT LIABILITY BEYOND THE REMEDIES  
SET FORTH IN THIS LIMITED WARRANTY STATEMENT OR  
LIABILITY FOR INCIDENTAL, INDIRECT, SPECIAL, PUNITIVE, OR  
CONSEQUENTIAL DAMAGES, INCLUDING WITHOUT  
LIMITATION ANY LIABILITY FOR PRODUCTS NOT BEING  
AVAILABLE FOR USE OR FOR LOST DATA OR SOFTWARE.  
Appendix B: Warranties and Return Policy  
227  
SOME STATES (OR JURISDICTIONS) DO NOT ALLOW THE  
EXCLUSION OR LIMITATION OF INCIDENTAL, INDIRECT,  
SPECIAL, PUNITIVE, OR CONSEQUENTIAL DAMAGES, SO THE  
PRECEDING EXCLUSION OR LIMITATION MAY NOT APPLY TO  
YOU.  
These provisions apply to Dell’s one-year limited warranty only. For  
provisions of any service contract covering your system, see your invoice or  
the separate service contract that you will receive.  
If Dell elects to exchange a product or portion of a product, the exchange  
will be made in accordance with Dell’s Exchange Policy in effect on the date  
of the exchange. In any instance in which Dell issues a Return Material  
Authorization Number, Dell must receive the product(s) for repair prior to  
the expiration of the warranty period in order for the repair(s) to be covered  
by the limited warranty.  
Two-Year Limited Warranty (Canada Only)  
Dell Computer Corporation ("Dell") manufactures its hardware products  
from parts and components that are new or equivalent to new in accordance  
with industry-standard practices. Dell warrants that the hardware products  
it manufactures will be free from defects in materials and workmanship.  
The warranty term is two years beginning on the date of invoice, as  
described in the following text.  
Damage due to shipping the products to you is covered under this limited  
warranty. Otherwise, this limited warranty does not cover damage due to  
external causes, including accident, abuse, misuse, problems with electrical  
power, servicing not authorized by Dell, usage not in accordance with  
product instructions, failure to perform required preventive maintenance,  
and problems caused by use of parts and components not supplied by Dell.  
228  
Appendix B: Warranties and Return Policy  
This limited warranty does not cover any items that are in one or more of  
the following categories: software; external devices (except as specifically  
noted); accessories or parts added to a Dell system after the system is  
shipped from Dell; accessories or parts added to a Dell system through  
Dell’s system integration department; accessories or parts that are not  
installed in the Dell factory; or Dell Software and Peripherals products.  
Monitors, keyboards, and mice that are Dell-branded or that are included  
on Dell’s standard price list are covered under this limited warranty; all  
other monitors, keyboards, and mice (including those sold through the Dell  
Software and Peripherals program) are not covered. Batteries for portable  
computers are covered only during the initial one-year period of this limited  
warranty.  
Limited Warranty Coverage During Year One  
During the one-year period beginning on the invoice date, Dell will repair or  
replace products covered under this limited warranty that are returned to  
Dell's facility. To request warranty service, you must contact Dell’s  
Customer Technical Support within the warranty period. If warranty service  
is required, Dell will issue a Return Material Authorization Number. You  
must ship the products back to Dell in their original or equivalent  
packaging, prepay shipping charges, and insure the shipment or accept the  
risk of loss or damage during shipment. Dell will ship the repaired or  
replacement products to you freight prepaid if you use an address in  
Canada, where applicable. Shipments to other locations will be made  
freight collect.  
NOTE: Before you ship  
the product(s) to Dell,  
back up the data on the  
hard drive(s) and any  
other storage device(s) in  
the product(s). Remove  
any removable media,  
such as diskettes, CDs, or  
PC Cards. Dell does not  
accept liability for lost  
data or software.  
Dell owns all parts removed from repaired products. Dell uses new and  
reconditioned parts made by various manufacturers in performing warranty  
repairs and building replacement products. If Dell repairs or replaces a  
product, its warranty term is not extended.  
Appendix B: Warranties and Return Policy  
229  
Limited Warranty Coverage During Year Two  
During the second year of this limited warranty, Dell will provide, on an  
exchange basis and subject to Dell’s Exchange Policy in effect on the date of  
the exchange, replacement parts for the Dell hardware product(s) covered  
under this limited warranty when a part requires replacement. You must  
report each instance of hardware failure to Dell’s Customer Technical  
Support in advance to obtain Dell’s concurrence that a part should be  
replaced and to have Dell ship the replacement part. Dell will ship parts  
(freight prepaid) if you use an address in Canada, where applicable.  
Shipments to other locations will be made freight collect. Dell will include a  
prepaid shipping container with each replacement part for your use in  
returning the replaced part to Dell. Replacement parts are new or  
reconditioned. Dell may provide replacement parts made by various  
manufacturers when supplying parts to you. The warranty term for a  
replacement part is the remainder of the limited warranty term.  
You will pay Dell for replacement parts if the replaced part is not returned to  
Dell. The process for returning replaced parts, and your obligation to pay for  
replacement parts if you do not return the replaced parts to Dell, will be in  
accordance with Dell’s Exchange Policy in effect on the date of the  
exchange.  
You accept full responsibility for your software and data. Dell is not required  
to advise or remind you of appropriate backup and other procedures.  
General Provisions  
DELL MAKES NO EXPRESS WARRANTIES OR CONDITIONS  
BEYOND THOSE STATED IN THIS LIMITED WARRANTY  
STATEMENT. DELL DISCLAIMS ALL OTHER WARRANTIES AND  
CONDITIONS, EXPRESS OR IMPLIED, INCLUDING WITHOUT  
LIMITATION IMPLIED WARRANTIES AND CONDITIONS OF  
MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.  
SOME STATES (OR JURISDICTIONS) DO NOT ALLOW  
LIMITATIONS ON IMPLIED WARRANTIES OR CONDITIONS, SO  
THIS LIMITATION MAY NOT APPLY TO YOU.  
230  
Appendix B: Warranties and Return Policy  
DELL'S RESPONSIBILITY FOR MALFUNCTIONS AND DEFECTS IN  
HARDWARE IS LIMITED TO REPAIR AND REPLACEMENT AS SET  
FORTH IN THIS LIMITED WARRANTY STATEMENT. THESE  
WARRANTIES GIVE YOU SPECIFIC LEGAL RIGHTS, AND YOU MAY  
ALSO HAVE OTHER RIGHTS, WHICH VARY FROM STATE TO  
STATE (OR JURISDICTION TO JURISDICTION).  
DELL DOES NOT ACCEPT LIABILITY BEYOND THE REMEDIES  
SET FORTH IN THIS LIMITED WARRANTY STATEMENT OR  
LIABILITY FOR INCIDENTAL, INDIRECT, SPECIAL, PUNITIVE, OR  
CONSEQUENTIAL DAMAGES, INCLUDING WITHOUT  
LIMITATION ANY LIABILITY FOR PRODUCTS NOT BEING  
AVAILABLE FOR USE OR FOR LOST DATA OR SOFTWARE.  
SOME STATES (OR JURISDICTIONS) DO NOT ALLOW THE  
EXCLUSION OR LIMITATION OF INCIDENTAL, INDIRECT,  
SPECIAL, PUNITIVE, OR CONSEQUENTIAL DAMAGES, SO THE  
PRECEDING EXCLUSION OR LIMITATION MAY NOT APPLY TO  
YOU.  
These provisions apply to Dell’s two-year limited warranty only. For  
provisions of any service contract covering your system, see your invoice or  
the separate service contract that you will receive.  
If Dell elects to exchange a system or component, the exchange will be  
made in accordance with Dell’s Exchange Policy in effect on the date of the  
exchange. In any instance in which Dell issues a Return Material  
Authorization Number, Dell must receive the product(s) for repair prior to  
the expiration of the warranty period in order for the repair(s) to be covered  
by the limited warranty.  
Three-Year Limited Warranty (Canada Only)  
Dell Computer Corporation ("Dell") manufactures its hardware products  
from parts and components that are new or equivalent to new in accordance  
with industry-standard practices. Dell warrants that the hardware products  
it manufactures will be free from defects in materials and workmanship.  
The warranty term is three years beginning on the date of invoice, as  
described in the following text.  
Appendix B: Warranties and Return Policy  
231  
Damage due to shipping the products to you is covered under this limited  
warranty. Otherwise, this limited warranty does not cover damage due to  
external causes, including accident, abuse, misuse, problems with electrical  
power, servicing not authorized by Dell, usage not in accordance with  
product instructions, failure to perform required preventive maintenance,  
and problems caused by use of parts and components not supplied by Dell.  
This limited warranty does not cover any items that are in one or more of  
the following categories: software; external devices (except as specifically  
noted); accessories or parts added to a Dell system after the system is  
shipped from Dell; accessories or parts added to a Dell system through  
Dell’s system integration department; accessories or parts that are not  
installed in the Dell factory; or Dell Software and Peripherals products.  
Monitors, keyboards, and mice that are Dell-branded or that are included  
on Dell’s standard price list are covered under this limited warranty; all  
other monitors, keyboards, and mice (including those sold through the Dell  
Software and Peripherals program) are not covered. Batteries for portable  
computers are covered only during the initial one-year period of this limited  
warranty.  
Limited Warranty Coverage During Year One  
During the one-year period beginning on the invoice date, Dell will repair or  
replace products covered under this limited warranty that are returned to  
Dell’s facility. To request warranty service, you must contact Dell’s  
Customer Technical Support within the warranty period. To request  
warranty service, you must contact Dell’s Customer Technical Support  
within the warranty period. If warranty service is required, Dell will issue a  
Return Material Authorization Number. You must ship the products back to  
Dell in their original or equivalent packaging, prepay shipping charges, and  
insure the shipment or accept the risk of loss or damage during shipment.  
Dell will ship the repaired or replacement products to you freight prepaid if  
you use an address in Canada, where applicable. Shipments to other  
locations will be made freight collect.  
Before you ship the product(s) to Dell, back up the data on the hard drive(s)  
and any other storage device(s) in the product(s). Remove any removable  
media, such as diskettes, CDs, or PC Cards. Dell does not accept liability  
for lost data or software.  
232  
Appendix B: Warranties and Return Policy  
Dell owns all parts removed from repaired products. Dell uses new and  
reconditioned parts made by various manufacturers in performing warranty  
repairs and building replacement products. If Dell repairs or replaces a  
product, its warranty term is not extended.  
Limited Warranty Coverage During Years Two and Three  
During the second and third years of this limited warranty, Dell will provide,  
on an exchange basis and subject to Dell’s Exchange Policy in effect on the  
date of the exchange, replacement parts for the Dell hardware product(s)  
covered under this limited warranty when a part requires replacement. You  
must report each instance of hardware failure to Dell’s Customer Technical  
Support in advance to obtain Dell’s concurrence that a part should be  
replaced and to have Dell ship the replacement part. Dell will ship parts  
(freight prepaid) if you use an address in Canada, where applicable.  
Shipments to other locations will be made freight collect. Dell will include a  
prepaid shipping container with each replacement part for your use in  
returning the replaced part to Dell. Replacement parts are new or  
reconditioned. Dell may provide replacement parts made by various  
manufacturers when supplying parts to you. The warranty term for a  
replacement part is the remainder of the limited warranty term.  
These provisions apply to Dell’s three-year limited warranty only. For  
provisions of any service contract covering your system, see your invoice or  
the separate service contract that you will receive.  
If Dell elects to exchange a system or component, the exchange will be  
made in accordance with Dell’s Exchange Policy in effect on the date of the  
exchange. In any instance in which Dell issues a Return Material  
Authorization Number, Dell must receive the product(s) for repair prior to  
the expiration of the warranty period in order for the repair(s) to be covered  
by the limited warranty.  
Appendix B: Warranties and Return Policy  
233  
"Total Satisfaction" Return Policy (U.S. Only)  
If you are an end-user customer who bought new products directly from a  
Dell company, you may return them to Dell within 30 days of the date of  
invoice for a refund or credit of the product purchase price. If you are an  
end-user customer who bought reconditioned or refurbished products from  
a Dell company, you may return them to Dell within 14 days of the date of  
invoice for a refund or credit of the product purchase price. In either case,  
the refund or credit will not include any shipping and handling charges  
shown on your invoice. If you are an organization that bought the products  
under a written agreement with Dell, the agreement may contain different  
terms for the return of products than specified by this policy.  
To return products, you must contact Dell Customer Service to receive a  
Credit Return Authorization Number. To expedite the processing of your  
refund or credit, Dell expects you to return the products to Dell in their  
original packaging within five days of the date that Dell issues the Credit  
Return Authorization Number. You must also prepay shipping charges and  
insure the shipment or accept the risk of loss or damage during shipment.  
You may return software for refund or credit only if the sealed package  
containing the diskette(s) or CD(s) is unopened. Returned products must  
be in as-new condition, and all of the manuals, diskette(s), CD(s), power  
cables, and other items included with a product must be returned with it.  
For customers who want to return, for refund or credit only, either  
applications software or an operating system that has been installed by Dell,  
the whole system must be returned, along with any media and  
documentation that may have been included in the original shipment.  
This "Total Satisfaction" Return Policy does not apply to Dell Software and  
Peripherals products, which may be returned under Dell Software and  
Peripherals's then-current return policy. The "Total Satisfaction" Return  
Policy also does not apply to Dell | EMC storage products.  
234  
Appendix B: Warranties and Return Policy  
One-Year End-User Manufacturer Guarantee (Latin America and  
the Caribbean Only)  
Guarantee  
Dell Computer Corporation ("Dell") warrants to the end user in accordance  
with the following provisions that its branded hardware products, purchased  
by the end user from a Dell company or an authorized Dell distributor in  
Latin America or the Caribbean, will be free from defects in materials,  
workmanship, and design affecting normal use, for a period of one year from  
the original purchase date. Products for which proper claims are made will,  
at Dell’s option, be repaired or replaced at Dell’s expense. Dell owns all  
parts removed from repaired products. Dell uses new and reconditioned  
parts made by various manufacturers in performing repairs and building  
replacement products.  
Exclusions  
This Guarantee does not apply to defects resulting from: improper or  
inadequate installation, use, or maintenance; actions or modifications by  
unauthorized third parties or the end user; accidental or willful damage; or  
normal wear and tear.  
Making a Claim  
Claims must be made in Latin America or the Caribbean by contacting the  
Dell point of sale within the guarantee period. The end user must always  
supply proof of purchase, indicating name and address of the seller, date of  
purchase, model and serial number, name and address of the customer, and  
details of symptoms and configuration at the time of malfunction,  
including peripherals and software used. Otherwise, Dell may refuse the  
guarantee claim. Upon diagnosis of a warranted defect, Dell will make  
arrangements and pay for ground freight and insurance to and from Dell’s  
repair/replacement center. The end user must ensure that the defective  
product is available for collection properly packed in original or equally  
protective packaging together with the details listed above and the return  
number provided to the end user by Dell.  
Appendix B: Warranties and Return Policy  
235  
Limitation and Statutory Rights  
Dell makes no other warranty, guarantee or like statement other than as  
explicitly stated above, and this Guarantee is given in place of all other  
guarantees whatsoever, to the fullest extent permitted by law. In the absence  
of applicable legislation, this Guarantee will be the end user’s sole and  
exclusive remedy against Dell or any of its affiliates, and neither Dell nor  
any of its affiliates shall be liable for loss of profit or contracts, or any other  
indirect or consequential loss arising from negligence, breach of contract, or  
howsoever.  
This Guarantee does not impair or affect mandatory statutory rights of the  
end user against and/or any rights resulting from other contracts concluded  
by the end user with Dell and/or any other seller.  
Dell World Trade LP  
One Dell Way, Round Rock, TX 78682, USA  
Dell Computadores do Brasil Ltda (CNPJ No. 72.381.189/0001-10)/  
Dell Commercial do Brasil Ltda (CNPJ No. 03 405 822/0001-40)  
Avenida Industrial Belgraf, 400  
92990-000 - Eldorado do Sul – RS - Brasil  
Dell Computer de Chile Ltda  
Coyancura 2283, Piso 3- Of.302,  
Providencia, Santiago - Chile  
Dell Computer de Colombia Corporation  
Carrera 7 #115-33 Oficina 603  
Bogota, Colombia  
Dell Computer de Mexico SA de CV  
Paseo de la Reforma 2620 - 11° Piso  
Col. Lomas Altas  
11950 México, D.F.  
236  
Appendix B: Warranties and Return Policy  
Glossary  
Array—A grouping of hard disk drives combines the storage space on the  
hard drives into a single segment of contiguous storage space. The RAID  
controller can group hard drives on one or more channels into an array. A  
hot spare drive does not participate in an array.  
Array Management Software—Software that provides common control and  
management for a disk array. Array Management Software most often  
executes in a disk controller or intelligent host bus adapter, but it can also  
execute in a host computer. When it executes in a disk controller or adapter,  
Array Management Software is often called firmware.  
Array Spanning—Array spanning by a logical drive combines storage space  
in two arrays of hard drives into a single, contiguous storage space in a  
logical drive. The logical drive can span consecutively numbered arrays,  
each having the same number of hard drives. Array spanning promotes  
RAID level 1 to RAID level 10. See also Disk Spanning, and Spanning.  
Asynchronous Operations—Operations that are not related to each other in  
time and can overlap. The concept of asynchronous I/O operations is central  
to independent access arrays in throughput-intensive applications.  
ATA —(Advanced Technology Attachment) Also known at IDE (Integrated  
Device Electronics), this is a type of hard disk drive interface in which the  
controller electronics are integrated onto the drive itself. A separate adapter  
card is no longer needed, which reduces interface costs and makes it easier  
to implement firmware.  
Glossary  
237  
Cache Input/Output (I/O)—A small amount of fast memory that holds  
recently accessed data. Caching speeds subsequent access to the same data.  
It is most often applied to processor-memory access, but can also be used to  
store a copy of data accessible over a network. When data is read from or  
written to main memory, a copy is also saved in cache memory with the  
associated main memory address. The cache memory software monitors the  
addresses of subsequent reads to see if the required data is already stored in  
cache memory. If it is already in cache memory (a cache hit), it is read from  
cache memory immediately and the main memory read is aborted (or not  
started.) If the data is not cached (a cache miss), it is fetched from main  
memory and saved in cache memory.  
Channel—An electrical path for the transfer of data and control  
information between a disk and a disk controller.  
Consistency Check—An examination of the disk system to determine  
whether all conditions are valid for the specified configuration (such as  
parity).  
Cold Swap—A cold swap requires that you turn the power off before  
replacing defective hard drive in a disk subsystem.  
Data Transfer Capacity—The amount of data per unit time moved through  
a channel. For disk I/O, bandwidth is expressed in megabytes per second  
(MB/s).  
Degraded Drive—A disk drive that has become non-functional or has  
decreased in performance.  
Disk—A non-volatile, randomly addressable, rewritable mass storage device,  
including both rotating magnetic and optical disks and solid-state disks, or  
non-volatile electronic storage elements. It does not include specialized  
devices such as write-once-read-many (WORM) optical disks, nor does it  
include so-called RAM disks implemented using software to control a  
dedicated portion of a host computer’s volatile random access memory.  
Disk Array—A collection of disks from one or more disk subsystems  
combined with array management software. It controls the disks and  
presents them to the array operating environment as one or more virtual  
disks.  
238  
Glossary  
Disk Duplexing—A variation on disk mirroring in which a second disk  
adapter or host adapter and redundant hard drives are present.  
Disk Mirroring—Writing duplicate data to more than one (usually two)  
hard disks to protect against data loss in the event of device failure. Disk  
mirroring is a common feature of RAID systems.  
Disk Spanning—Disk spanning allows multiple hard drives to function like  
one big drive. Spanning overcomes lack of disk space and simplifies storage  
management by combining existing resources or adding relatively  
inexpensive resources. For example, four 400 MB hard drives can be  
combined to appear to the operating system as one single 1600 MB drive.  
See also Array Spanning and Spanning.  
Disk Striping—A type of disk array mapping. Consecutive stripes of data  
are mapped round-robin to consecutive array members. A striped array  
(RAID level 0) provides high I/O performance at low cost, but provides  
lowers data reliability than any of its member disks.  
Disk Subsystem—A collection of disks and the hardware that connects  
them to one or more host computers. The hardware can include an  
intelligent controller, or the disks can attach directly to a host computer.  
Double Buffering—A technique that achieves maximum data transfer  
bandwidth by constantly keeping two I/O requests for adjacent data  
outstanding. A software component begins a double-buffered I/O stream by  
issuing two requests in rapid sequence. Thereafter, each time an I/O request  
completes, another is immediately issued. If the disk subsystem is capable  
of processing requests fast enough, double buffering allows data to be  
transferred at the full-volume transfer rate.  
Failed Drive—A drive that has ceased to function or consistently functions  
improperly.  
Firmware—Software stored in read-only memory (ROM) or Programmable  
ROM (PROM). Firmware is often responsible for the behavior of a system  
when it is first turned on. A typical example would be a monitor program in  
a computer that loads the full operating system from disk or from a network  
and then passes control to the operating system.  
Glossary  
239  
FlexRAID Power Fail Option—The FlexRAID Power Fail option allows a  
reconstruction to restart if a power failure occurs. This is the advantage of  
this option. The disadvantage is, once the reconstruction is active, the  
performance is slower because an additional activity is running.  
Formatting—The process of writing zeros to all data fields in a physical  
drive (hard drive) to map out unreadable or bad sectors. Because most hard  
drives are factory formatted, formatting is usually only done if a hard disk  
generates many media errors.  
GB—(gigabyte) 1,000,000,000 (10 to the ninth power) bytes. It is the same  
as 1,000 MB (megabytes).  
Host-based Array—A disk array with an Array Management Software in its  
host computer rather than in a disk subsystem.  
Host Computer—Any computer to which disks are directly attached.  
Mainframes, servers, workstations, and personal computers can all be  
considered host computers.  
Hot Spare—A stand-by drive ready for use if another drive fails. It does not  
contain any user data. Up to eight hard drives can be assigned as hot spares  
for an adapter. A hot spare can be dedicated to a single redundant array or it  
can be part of the global hot-spare pool for all arrays controlled by the  
adapter.  
Hot Swap—The substitution of a replacement unit in a disk subsystem for  
a defective one, where the substitution can be performed while the  
subsystem is running (performing its normal functions). Hot swaps are  
manual.  
Input/Output (I/O) Driver—A host computer software component  
(usually part of the operating system) that controls the operation of  
peripheral controllers or adapters attached to the host computer. I/O drivers  
communicate between applications and I/O devices, and in some cases  
participates in data transfer.  
Integrated Device Electronics (IDE)— Also known at Advanced  
Technology Attachment (ATA), this is a type of interface for the hard drive,  
in which the controller electronics are integrated onto the drive itself. With  
IDE, a separate adapter card is no longer needed; this reduces interface  
costs and makes it easier to implement firmware.  
240  
Glossary  
Initialization—The process of writing zeros to the data fields of a logical  
drive and generating the corresponding parity to bring the logical drive to a  
ready state. Initializing erases previous data and generates parity so that the  
logical drive will pass a consistency check. Arrays can work without  
initializing, but they can fail a consistency check because the parity fields  
have not been generated.  
Logical Disk—A set of contiguous chunks on a physical disk. Logical disks  
are used in array implementations as constituents of logical volumes or  
partitions. Logical disks are normally transparent to the host environment,  
except when the array containing them is being configured.  
Logical Drive—A virtual drive within an array that can consist of more than  
one physical drive. Logical drives divide the contiguous storage space of an  
array of hard drives or a spanned group of arrays of drives. The storage space  
in a logical drive is spread across all the physical drives in the array or  
spanned arrays. Configure at least one logical drive for each array.  
Mapping—The conversion between multiple data addressing schemes,  
especially conversions between member disk block addresses and block  
addresses of the virtual disks presented to the operating environment by  
Array Management Software.  
Megabyte (MB)— An abbreviation for 1,000,000 (10 to the sixth power)  
bytes. It is the same as 1,000 KB (kilobytes).  
Multi-threaded—Having multiple concurrent or pseudo-concurrent  
execution sequences. Used to describe processes in computer systems.  
Multi-threaded processes allow throughput-intensive applications to  
efficiently use a disk array to increase I/O performance.  
Operating Environment—The operating environment includes the host  
computer where the array is attached, any I/O buses and adapters, the host  
operating system, and any additional software required to operate the array.  
For host-based arrays, the operating environment includes I/O driver  
software for the member disks, but does not include Array Management  
Software, which is regarded as part of the array itself.  
Glossary  
241  
Parity—Parity is an extra bit added to a byte or word to reveal errors in  
storage (in random access memory, RAM, or disk) or transmission. Parity is  
used to generate a set of redundancy data from two or more parent data  
sets. The redundancy data can be used to reconstruct one of the parent data  
sets. However, parity data does not fully duplicate the parent data sets. In  
RAID, this method is applied to entire drives or stripes across all hard drives  
in an array. Parity consists of dedicated parity, in which the parity of the data  
on two or more drives is stored on an additional drive, and distributed parity,  
in which the parity data are distributed among all the drives in the system.  
If a single drive fails, it can be rebuilt from the parity of the respective data  
on the remaining drives.  
Partition—An array virtual disk made up of logical disks rather than  
physical ones. Also known as logical volume.  
Physical Drive—A hard drive that stores data. A hard drive consists of one  
or more rigid magnetic discs rotating about a central axle with associated  
read/write heads and electronics.  
Physical Drive Roaming—The ability of some adapters to detect when hard  
drives have been moved to a different slots in the computer, for example,  
after a hot swap.  
Protocol—A set of formal rules describing how to transmit data, especially  
across a network. Low level protocols define the electrical and physical  
standards to be observed, bit- and byte- ordering, and the transmission and  
error detection and correction of the bit stream. High level protocols deal  
with the data formatting, including the message syntax, the terminal-to-  
computer dialogue, character sets, and sequencing of messages.  
RAID—Redundant Array of Independent Disks. An array of multiple  
independent hard disk drives that yields better performance than a Single  
Large Expensive Disk (SLED). A RAID disk subsystem improves I/O  
performance on a server using only a single drive. The RAID array appears to  
the host server as a single storage unit. I/O is expedited because several disks  
can be accessed simultaneously.  
242  
Glossary  
RAID Levels—A style of redundancy applied to a logical drive. It can  
increase the performance of the logical drive and can decrease usable  
capacity. Each logical drive must have a RAID level assigned to it. The  
RAID level drive requirements are: RAID 0 requires one to eight physical  
drives, RAID 1 requires exactly two physical drives, RAID 5 requires three to  
eight physical drives and RAID 10 requires four, six or eight physical drives.  
RAID 10 results when a RAID 1 logical drive spans arrays.  
RAID Migration—RAID migration is used to move between optimal RAID  
levels or to change from a degraded redundant logical drive to an optimal  
RAID 0. In Novell, the utility used for RAID migration is MEGAMGR. If a  
RAID 1 is being converted to a RAID 0, instead of performing RAID  
migration, one drive can be removed and the other reconfigured on the  
controller as a RAID 0. This is due to the same data being written to each  
drive.  
Read-Ahead—A memory caching capability in some adapters that allows  
them to read sequentially ahead of requested data and store the additional  
data in cache memory, anticipating that the additional data will be needed  
soon. Read-Ahead supplies sequential data faster, but is not as effective  
when accessing random data.  
Ready State—A condition in which a workable hard drive is neither online  
nor a hot spare and is available to add to an array or to designate as a hot  
spare.  
Rebuild—The regeneration of all data from a failed disk in a RAID level 1,  
or 5 array to a replacement disk. A disk rebuild normally occurs without  
interruption of application access to data stored on the array virtual disk.  
Rebuild Rate—The percentage of CPU resources devoted to rebuilding.  
Reconstruct—The act of remaking a logical drive after changing RAID  
levels or adding a physical drive to an existing array.  
Redundancy—The provision of multiple interchangeable components to  
perform a single function to cope with failures or errors. Redundancy  
normally applies to hardware; a common form of hardware redundancy is  
disk mirroring.  
Replacement Disk—A disk available to replace a failed member disk in a  
RAID array.  
Glossary  
243  
Replacement Unit—A component or collection of components in a disk  
subsystem that are always replaced as a unit when any part of the collection  
fails. Typical replacement units in a disk subsystem includes disks, controller  
logic boards, power supplies, and cables. Also called a hot spare.  
SCSI—Small Computer System Interface. A processor-independent  
standard for system-level interfacing between a computer and intelligent  
devices, including hard disks, diskettes, CD drives, printers, scanners, etc.  
SCSI can connect up to seven devices to a single adapter (or host adapter)  
on the computer's bus. SCSI transfers eight or 16 bits in parallel and can  
operate in either asynchronous or synchronous modes. The synchronous  
transfer rate is up to 160 MB/s. SCSI connections normally use single ended  
drivers, as opposed to differential drivers.  
The original standard is now called SCSI-1 to distinguish it from SCSI-2  
and SCSI-3, which include specifications of Wide SCSI (a 16-bit bus) and  
Fast SCSI (10 MB/s transfer.) Ultra 160M SCSI is a subset of Ultra3 SCSI  
and allows a maximum throughput of 160 MB/s, which is more than twice  
as fast as Wide Ultra2 SCSI.  
Service Provider—The Service Provider (SP) is a program that resides in the  
desktop system or server and is responsible for all Desktop Management  
Interface (DMI) activities. This layer collects management information  
from products (whether system hardware, peripherals or software) stores  
that information in the DMI database and passes it to management  
applications as requested.  
SNMP—Simple Network Management Protocol, the most widely used  
protocol for communication management information between the  
managed elements of a network and a network manager. SNMP focuses  
primarily on the network backbone. The Internet standard protocol that  
manages nodes on an Internet Protocol (IP) network.  
Spanning—Array spanning by a logical drive combines storage space in two  
arrays of hard drives into a single, contiguous storage space in a logical drive.  
Logical drives can span consecutively numbered arrays that each consist of  
the same number of hard drives. Array spanning promotes RAID level 1 to  
RAID levels 10. See also Array Spanning, and Disk Spanning.  
Spare—A hard drive available to back up the data of other drives.  
244  
Glossary  
Stripe Size—The amount of data contiguously written to each disk. You can  
specify stripe sizes of 4 KB, 8 KB, 16 KB, 32 KB, 64 KB, and 128 KB for each  
logical drive. For best performance, choose a stripe size equal to or smaller  
than the block size used by the host computer.  
Stripe Width—The number of hard drives across which the data are striped.  
Striping—Segmentation of logically sequential data, such as a single file, so  
that segments can be written to multiple physical devices in a round-robin  
fashion. This technique is useful if the processor can read or write data  
faster than a single disk can supply or accept it. While data is being  
transferred from the first disk, the second disk can locate the next segment.  
Data striping is used in some modern databases and in certain RAID  
devices.  
Stripe Size—The amount of data contiguously written to each disk. Also  
called "stripe depth." You can specify stripe sizes of 4 KB, 8 KB, 16 KB, 32  
KB, 64 KB, and 128 KB for each logical drive. A larger stripe size produces  
improved read performance, especially if most of the reads are sequential.  
For mostly random reads, select a smaller stripe size.  
Stripe Width—The number of drives across which data is striped.  
Terminator—A resistor connected to a signal wire in a bus or network for  
impedance matching to prevent reflections, e.g., a resistor connected across  
signal wires at the end of a SCSI cable.  
Wide SCSI—A variant on the SCSI-2 interface. Wide SCSI has a 16-bit  
bus, double the width of the original SCSI-1.  
Write-Back—In Write-Back caching mode, the controller sends a data  
transfer completion signal to the host when the controller cache has  
received all the data in a disk write transaction. Data are written to the disk  
subsystem in accordance with policies set up by the controller. These  
policies include the amount of dirty/clean cache lines, the number of cache  
lines available, elapsed time from the last cache flush, and others.  
Write-Through—In Write-Through caching, the controller sends a data  
transfer completion signal to the host when the disk subsystem has received  
all the data in a transaction. The controller cache is not used.  
Glossary  
245  
246  
Glossary  
Index  
BIOS Configuration  
Utility, 72, 81, 115-116,  
147, 149-159, 163, 167,  
173, 178-185, 188-193,  
199, 206, 208  
Check Consistency Menu, 159  
Configure Menu, 152  
Format Menu, 158  
Initialize Menu, 153  
Menu Options, 151, 153  
Objects Menu, 153  
Rebuild Menu, 158  
Starting, 150  
Cluster Mode, 154-155, 185-  
186  
Enabling and Disabling, 185  
Initiator ID, 154-155, 186  
A
Array, 237  
Array Configuration  
Planner, 98  
Cold Swap, 238  
Array Management  
Software, 237  
Configuration Features, 54-  
55, 64, 74  
Array Performance  
Features, 56, 67, 76  
Configuration Method, 160  
Easy Configuration, 96, 152,  
159-160, 163, 167, 183  
New Configuration, 96, 152,  
159-161, 167, 208  
Array Spanning, 237  
Arrays, 92-93, 109, 125-126,  
142, 170, 176  
View/Add Configuration, 150,  
152, 159, 161-162, 167,  
173, 206  
Asynchronous  
BIOS Error Messages,  
Other, 208  
Operations, 237  
ATA, 237  
Configuration on Disk, 55,  
65, 75, 150, 161, 208  
Audible Warnings, 210  
C
Configuration Strategies, 93  
Maximizing Capacity, 94  
Maximizing Drive  
Availability, 94  
Maximizing Drive  
Cables  
SCSI, 102, 104, 113-114,  
118, 128, 134, 136, 146,  
204  
B
Battery Backup, 193  
BIOS, 58, 60, 68, 70, 72, 78,  
80-81, 93, 96, 104-105,  
115-116, 121-122, 130,  
136-137, 139-140, 146-  
147, 149-159, 163, 167,  
173, 178-185, 188-193,  
199, 204-208  
Cache I/O, 238  
Performance, 95  
Cache Memory, 59, 70, 79  
Configuring Arrays and  
Logical Drives, 159  
Cache Policy, 87, 161, 166,  
172, 177, 192, 197  
Consistency Check, 32, 238  
Channel, 238  
Cluster, 65, 154-155, 161,  
166, 185-186  
BIOS Boot Error  
Messages, 205  
D
Data Storage, 96  
Index  
247  
Optimizing, 96  
Firmware, 239  
I
Data Transfer Capacity, 238  
Degraded, 238  
FlexRAID Power Fail  
Option, 240  
I/O Driver, 240  
IDE, 240  
Format, 240  
Dell Manager, 61-62, 72, 81-  
82, 187-190, 194-195,  
197-202  
Check Consistency, 195  
Format Menu, 194  
Initialize Menu, 190  
Objects Menu, 190  
Rebuild Menu, 195  
Starting, 188  
Initialization, 153, 179, 241  
Batch, 178, 197  
Individual, 178-179, 197-198  
Logical Drives, 178-180, 199  
Formatting, 100, 158, 179-  
181, 194, 198-200  
Batch, 180-181, 200  
Individual, 181, 200  
Media Errors, 180, 199  
Physical Drives, 180, 199  
J
Jumpers Setting  
J10 TERMPWR Enable, 105,  
107, 121, 123, 137-138  
J11 TERMPWR Enable, 105,  
113, 121, 123-124, 137-  
138  
J12 TERMPWR Enable, 137-  
138, 140  
J14 Serial Port, 122, 124, 137,  
139  
G
Disk, 238  
GB, 240  
Disk Array, 238  
Disk Duplexing, 239  
Disk Mirroring, 34, 239  
Disk Spanning, 35, 239  
Disk Striping, 32, 239  
Disk Subsystem, 239  
Double Buffering, 239  
H
Hardware Architecture  
Features, 56, 66, 76  
Hardware Installation, 101,  
117, 133  
J17 Dirty Cache LED, 105,  
108, 122, 125, 139  
J19 Onboard BIOS  
Installation Steps, 103, 136  
Enable, 137, 139  
Hardware Requirements, 54,  
64, 74  
J2 Termination Enable, 105,  
121-122, 125, 137-138,  
184  
J23 External Battery, 137, 140  
J3 Termination Enable, 105,  
121-122, 125, 137-138,  
184  
J5 Termination Enable, 105-  
106, 121, 128, 137-138,  
184  
E
Hardware Termination, 184  
Host Computer, 240  
Easy Configuration, 96, 152,  
159-160, 163, 167, 183  
Enclosure Management, 40  
Host-based Array, 240  
Hot Spares, 37, 162, 195, 240  
Creating, 93  
Designating Drives as, 162,  
195  
F
Failed Drive, 239  
J7 Termination Enable, 105,  
121, 128, 137-138, 184  
J9 TERMPWR Enable, 105-  
106, 121-122, 137-138  
Hot Swap, 39, 240  
Fault Tolerance, 31, 57, 67,  
77, 94, 98  
Features, 57, 67, 77  
248  
Index  
Partition, 242  
171, 174-176, 180, 182,  
188-189, 191, 195-196,  
201, 209-210, 242  
L
PCI Bridge, 59, 69, 79  
Logical Disk, 241  
PERC, 24-27, 30-33, 37-40,  
42-43, 45, 47, 53-84, 92-  
93, 96-97, 101-105, 108-  
109, 111-113, 115-122,  
125-128, 130-131, 133-  
137, 139-141, 143-144,  
146-147, 149-155, 157,  
159, 161-163, 166-168,  
170, 173, 176, 178-185,  
189-191, 195, 199, 201,  
204-210  
Logical Drives, 115, 130, 147,  
150, 159, 205, 207, 241  
Configuring, 96  
Creating, 93  
Definition, 183  
Initializing, 178, 197  
Spanning, 35  
States, 40  
Levels, 31, 41, 54, 59, 64, 69,  
74, 79, 98  
Management, 61, 72, 81  
Overview, 31  
RAID 0, 36, 42-43, 48-50, 94,  
98-99, 164, 168, 174, 210  
RAID 1, 34-36, 39, 45, 48-49,  
98-99, 164, 166-168, 170-  
171, 174, 176, 182, 196,  
201, 210  
BIOS, 58, 60-61, 72  
Features, 53  
M
RAID 10, 35-36, 48-49, 98-99,  
168, 170, 174, 176  
Mapping, 241  
MB, 241  
PERC Configuration  
Manager, 93, 96, 206-207,  
209  
RAID 5, 36, 42, 46-47, 50-51,  
92, 94, 98-99, 164, 168-  
170, 174, 176, 210  
Media Errors, 180, 199  
Multi-threaded, 241  
Physical Disk, 242  
RAID 50, 36, 50-51, 92, 98-99,  
169-170, 174, 176  
Physical Disk Roaming, 242  
Physical Drives, 84, 87, 95,  
180  
Formatting, 180, 199  
Media Errors, 180, 199  
RAID Levels, 31, 41, 54, 59,  
64, 69, 74, 79, 98, 243  
Assigning, 95  
N
New Configuration, 96, 152,  
159-161, 167, 208  
RAID Migration, 243  
Read Policy, 87, 192, 196  
Read-Ahead, 243  
Pre-loaded SCSI Drive  
Using As-Is, 183  
O
Protocol, 242  
Operating Environment, 241  
Ready State, 243  
Operating System Software  
Drivers, 58, 68, 78  
Rebuild Rate, 243  
R
Rebuilding, 39, 182-183, 201-  
202  
Failed Disk Drives, 182, 201  
Rebuild Types, 182, 201  
RAID, 24-25, 29-32, 34-39,  
41-43, 45-51, 54-55, 58-  
59, 61-62, 64-65, 69, 71-  
72, 74-75, 78-79, 81-82,  
84, 87, 92-99, 108, 115,  
123, 130, 146, 150, 152,  
155, 159-160, 162, 164-  
P
Parameters, 155, 191, 196,  
209  
Reconstruction, 243  
Redundancy, 243  
Parity, 36, 57, 67, 77, 242  
Replacement Disk, 243  
249  
Index  
Replacement Unit, 244  
Spanning, 244  
Spare, 244  
Write Policy, 87, 161, 166,  
171, 176, 191, 196  
Return Policy, 213, 234  
Stripe Size, 33, 57, 61, 67, 71,  
81, 87, 161, 166, 196, 245  
S
Stripe Width, 245  
Striping, 245  
Safety Instructions, 1  
SCSI, 24-27, 30-31, 39, 54,  
56, 59-61, 64, 66, 69-71,  
74, 76, 79-81, 84-87, 89,  
102-109, 113-115, 118,  
120-122, 124-126, 128-  
130, 134, 136-138, 140-  
142, 145-147, 153, 156-  
158, 166, 171, 177, 180,  
183-184, 190, 193-194,  
196, 199, 204-205, 207-  
208, 211, 244  
Bus, 26, 59-61, 69-71, 79-81  
Bus Widths and Maximum  
Throughput, 26  
Cables, 113, 129, 145, 211  
Channel, 25, 27, 54, 59, 64,  
69, 74, 79, 84-87, 89,  
122, 138, 157, 193  
T
Termination, 59-60, 69, 71,  
79-80, 105, 108, 121-122,  
125-126, 137-138, 140,  
157, 184, 193  
Hardware, 184  
SCSI, 59-60, 69, 71, 79-80,  
105, 108-109, 122, 125,  
138, 141  
Troubleshooting  
BIOS Boot Error  
Messages, 205  
V
Connectors, 25, 59-60, 69,  
71, 79-80  
Firmware, 61, 71, 81  
Termination, 59-60, 69, 71,  
79-80, 105, 108-109,  
122, 125, 138, 141  
View/Add Configuration, 150,  
152, 159, 161-162, 167,  
173, 206  
Virtual Sizing, 55, 65, 75, 155,  
161, 191, 209  
Serial Port, 59-60, 69-70, 79-  
80, 106, 123-124, 139  
W
Service Provider, 244  
Warranties, 213  
SMART Technology, 55, 65,  
75  
WebBIOS Configuration  
Utility, 61-62, 72, 81-82,  
104, 116, 130, 136, 147  
SNMP, 244  
Software Utilities, 77  
250  
Index  
Printed in the U.S.A.  
P/N 5C229 Rev. A04  
w w w. d e l l . c o m | s u p p o r t . d e l l . c o m  

HP Hewlett Packard L1740 PL766AA User Manual
HP Hewlett Packard Hewlett Packard Computer Keyboard TFT7600 User Manual
Epson C82605 User Manual
Electrolux Thermaline 9CHG584155 User Manual
Dell Computer Monitor IN2010N User Manual
Cypress CY7C1910BV18 User Manual
Black Decker TRO4070DC User Manual
Black Box Computer Hardware Dual Link DVI Splitters User Manual
Asus Z87 Plus Motherboard Z87PLUS User Manual
Asus MOTHERBOARD P4P800S User Manual