HP Hewlett Packard Network Card HP IO User Manual

HP IO Accelerator Version 3.2.3 Windows  
User Guide  
Abstract  
This document describes software requirements for all relevant HP IO Accelerators using Microsoft Windows operating systems. This document is  
intended for system administrators who plan to install and use HP IO Accelerators with a Microsoft Windows operating system. It is helpful to have  
previous experience with HP IO Accelerators and a Microsoft Windows operating system. This user guide is intended for IO Accelerator software  
release 3.2.3 or later.  
Part Number: 647099-003  
February 2013  
Edition: 3  
Download from Www.Somanuals.com. All Manuals Search And Download.  
Contents  
Contents  
3
Download from Www.Somanuals.com. All Manuals Search And Download.  
Contents  
4
Download from Www.Somanuals.com. All Manuals Search And Download.  
Contents  
5
Download from Www.Somanuals.com. All Manuals Search And Download.  
About this guide  
Contents summary  
Instructions on downloading and installing the approved driver and utilities  
Instructions on maintaining the IO Accelerator  
Description of the following IO Accelerator models:  
o
o
o
HP IO Accelerator for BladeSystem c-Class  
HP PCIe IO Accelerator  
HP PCIe IO Accelerator Duo  
CAUTION: Before upgrading to 3.x.x software and firmware, back up all data on the IO  
Accelerator. The 3.2.3 software and firmware reformat the drive, which causes data to be lost if  
not backed up. The 3.2.3 software is not backward compatible with 1.2.x or 2.x software.  
About this guide  
6
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Introduction  
Overview  
Designed around ioMemory, a revolutionary storage architecture, HP IO Accelerator is an advanced NAND  
flash storage device. With performance comparable to DRAM and storage capacity on par with hard disks,  
the IO Accelerator increases performance so that every server can contain internal storage that exceeds the  
I/O performance of an enterprise SAN.  
HP IO Accelerator is the first data accelerator designed specifically to improve the bandwidth for I/O-bound  
applications.  
In addition to the hardware driver, the IO Accelerator also includes a VSL. This hybrid of the RAM  
virtualization subsystem and the disk I/O subsystem combines the best features of both systems. VSL functions  
as a disk to interface well with block-based applications and software, while also running like RAM  
underneath to maximize performance. This feature produces the following benefits:  
Performance: The VSL offers direct and parallel access to multiple CPU cores, enabling near linear  
performance scaling, consistent performance across different read/write workloads, and low latency  
with minimal interruptions and context switching.  
Extensibility: The VSL enables flash-optimized software development, making each IO Accelerator  
module a flexible building block for creating a flash-optimized data center.  
Product naming  
HP IO Accelerator Generation 1 devices include:  
AJ876A: HP 80GB IO Accelerator for BladeSystem c-Class  
AJ877A: HP 160GB IO Accelerator for BladeSystem c-Class  
AJ878A: HP 320GB IO Accelerator for BladeSystem c-Class  
AJ878B: HP 320 GB IO MLC Accelerator for BladeSystem c-Class  
BK836A: HP 640GB IO MLC Accelerator for BladeSystem c-Class  
IMPORTANT: Generation 1 IO accelerators for BladeSystem c-Class are only compatible with  
G7 and earlier server blades.  
600278-B21: HP 160GB Single Level Cell PCIe ioDrive for ProLiant Servers  
600279-B21: HP 320GB Multi Level Cell PCIe ioDrive for ProLiant Servers  
600281-B21: HP 320GB Single Level Cell PCIe ioDrive Duo for ProLiant Servers  
600282-B21: HP 640GB Multi Level Cell PCIe ioDrive Duo for ProLiant Servers  
641027-B21: HP 1.28TB Multi Level Cell PCIe ioDrive Duo for ProLiant Servers  
HP IO Accelerator Generation 2 devices include:  
QK761A: HP 365GB IO MLC Accelerator for BladeSystem c-Class  
Introduction  
7
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
QK762A: HP 785GB IO MLC Accelerator for BladeSystem c-Class  
QK763A: HP 1.2 TB IO MLC Accelerator for BladeSystem c-Class  
IMPORTANT: Generation 2 IO accelerators for BladeSystem c-Class are only compatible with  
Gen8 and later server blades.  
673642-B21: HP 365GB Multi Level Cell G2 PCIe ioDrive2 for ProLiant Servers  
673644-B21: HP 785GB Multi Level Cell G2 PCIe ioDrive2 for ProLiant Servers  
673646-B21: HP 1205GB Multi Level Cell G2 PCIe ioDrive2 for ProLiant Servers  
673648-B21: HP 2410GB Multi Level Cell G2 PCIe ioDrive2 Duo for ProLiant Servers  
721458-B21: HP 3.0TB Multi Level Cell G2 PCIe ioDrive2 for ProLiant Servers  
Performance attributes  
IO Accelerator capacity  
Models AJ878B and  
BK836A  
320GB  
640GB  
MLC (Multi Level Cell)  
735 MB/s  
510 MB/s  
100,000  
MLC (Multi Level Cell)  
750 MB/s  
NAND type  
Read Bandwidth (64kB)  
Write Bandwidth (64kB)  
Read IOPS (512 Byte)  
Write IOPS (512 Byte)  
Mixed IOPS* (75/25 r/w)  
Access Latency (512 Byte)  
Bus Interface  
550 MB/s  
93,000  
141,000  
145,000  
67,000  
74,000  
30 µs  
30 µs  
PCI-Express x4  
PCI-Express Gen1 x4  
IO Accelerator capacity  
Models QK762A and  
QK763A  
785GB  
1.2TB  
MLC (Multi Level Cell)  
1.5 GB/s  
MLC (Multi Level Cell)  
1.5 GB/s  
NAND type  
Read Bandwidth (1MB)  
Write Bandwidth (1MB)  
Read IOPS (Seq. 512 Byte)  
Write IOPS (Seq. 512 Byte)  
1.1 GB/s  
1.3 GB/s  
443,000  
443,000  
530,000  
530,000  
141,000  
475,000  
68µs  
143,000  
475,000  
68µs  
Read IOPS (Rand. 512 Byte)  
Write IOPS (Rand. 512 Byte)  
Read Access Latency  
Write Access Latency  
Bus Interface  
15µs  
15µs  
PCI-Express Gen2 x4  
PCI-Express Gen2 x4  
*Performance achieved using multiprocessor enterprise server  
Enterprise data integrity  
Field upgradeability  
Green footprint, 7.5W nominal per device  
Introduction  
8
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
NOTE: MSI was disabled to obtain these statistics.  
Required operating environment  
The IO Accelerator is supported for use in the following operating environments:  
Windows Server 2003 (x86_64-bit only) SP2 (not supported on Gen8 servers)  
Windows Server 2008 (x86_64-bit only) R1 with SP2 or higher  
Windows Server 2008 (x86_64-bit only) R2  
Windows Server 2012 (x86_64-bit only)  
IMPORTANT: All operating systems must be 64-bit architecture.  
NOTE: IO Accelerators cannot be used as hibernation devices.  
Supported firmware revisions  
After February 19, 2013, all IO Accelerators ship with firmware version 7.1.13.109322 or higher. This  
firmware version only works with VSL 3.2.2 or higher. If you are installing a recently purchased or a  
replacement IO Accelerator into a system that already has IO Accelerators installed, then you must upgrade  
the firmware on the previously installed devices to 7.1.13.109322 or higher. You must also upgrade the VSL  
to 3.2.2 or higher. Upgrading the firmware and VSL on cards that were running firmware versions 6.x.x or  
higher and VSL 3.x.x or higher is NOT data destructive. However, HP recommends that you back up any  
data on the device before performing the upgrade. The latest supported version of the firmware and VSL can  
be found on the HP website (http://www.hp.com).  
Release  
Firmware revision  
17350  
1.2.4  
1.2.7  
1.2.8.4  
2.2.x  
2.2.3  
2.3.1  
3.1.1  
3.2.3  
36867 or 42014  
43246  
43674  
101583  
101971 4 or 101971_6  
107004 or greater  
109322  
Supported hardware  
HP IO Accelerator for BladeSystem c-Class  
BladeSystem c-Class IO Accelerators have two distinct designs for the respective server product lines. The G1  
through G7 IO Accelerator adapter is provided in a c-Class Type 1 Mezzanine card form factor. It can be  
installed in both Type 1 and Type 2 mezzanine slots within the c-Class blade G1 through G7 servers,  
Introduction  
9
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
enabling a total of two cards in a half-height server blade, and three cards in a full-height server blade and  
up to 6 in a double-high, double-wide server (BL680c).  
The Gen8 adapter is provided in a c-Class Type B Mezzanine card form factor. It can only be installed in  
Type B mezzanine slots within the Gen 8 or later servers, enabling one IO Accelerator in a half-height Gen8  
server.  
The Type I mezz card and the Type B mezz card are distinguished by the mezzanine connector. The type B  
card is slightly larger than a Type I mezz card.  
The amount of free RAM required by the driver depends on the size of the blocks used when writing to the  
drive. The smaller the blocks, the more RAM is required. The table below lists the guidelines for each 80GB  
of storage. For the latest information, see the QuickSpecs sheet to the HP IO Accelerator for HP BladeSystem  
c-Class at HP Customer Support (http://www.hp.com/support).  
The Remote Power Cut Module for the c-Class blade mezzanine card provides a higher level of protection in  
the event of a catastrophic power loss (for example, a user accidentally pulls the wrong server blade out of  
the slot). The Remote Power Cut Module ensures in-flight writes are completed to NAND flash in these  
catastrophic scenarios. Without the Remote Power Cut Module, write performance is slower. Writes are not  
acknowledged until the data is written to the NAND module, thereby slowing performance. When the  
Remote Power Cut Module is installed, writes are acknowledged by the IO Accelerator controller to the  
driver. The IO Accelerator controller then completes the write to the NAND module.  
The IO Accelerators (QK761A, QK762A, and QK763A) for Gen 8 BladeSystem c-Class have the power cut  
functionality embedded on the card. They offer the protection without requiring the remote power cut module.  
NOTE: The Remote Power Cut Module is used only in the AJ878B and BK836A models. Without  
the Remote Power Cut Module, write performance is slower.  
HP PCIe IO Accelerator minimum requirements  
An open PCI-Express slot—The accelerator requires a minimum of one half-length, half-height slot with  
a x4 physical connector. All four lanes must be connected for full performance. HP PCIe IO Accelerator  
Duo requires a minimum of a full-height, half-length slot with an x8 physical connection. If your system  
is using PCI 1.1, all x8 signaling lanes must be connected for full performance. If your system is using  
PCI 2.0, for full performance you only have to connect x4 signaling lanes.  
NOTE: For PCIe IO Accelerators, using PCIe slots greater than x4 does not improve  
performance.  
NOTE: The power cut feature is built into PCIe IO Accelerators; therefore, no Remote Power Cut  
Module is necessary.  
300 LFM of airflow at no greater than 50°C. To protect against thermal damage, the IO Accelerator  
also monitors the junction temperature of its controller. The temperature represents the internal  
temperature of the controller, and it is reported in fio-statusreport.The IO Accelerator begins  
throttling write performance when the junction temperature reaches 78°C. If the junction temperature  
continues to rise, the IO Accelerator shuts down when the temperature reaches 85°C.  
NOTE: If you experience write performance throttling due to high temperatures, see your  
computer documentation for details on increasing airflow, including fan speeds.  
Sufficient RAM to operate—The amount of RAM that the driver requires to manage the NAND flash  
varies according to the block size you select when formatting the device (filesystem format, not low-level  
format). For a virtual machine using an IO Accelerator directly (using PCI pass-through), consult the user  
guide for the installed operating system. The following table lists the amount of RAM required per  
Introduction 10  
Download from Www.Somanuals.com. All Manuals Search And Download.  
100GB of storage space, using various block sizes. The amount of RAM used in driver version 3.0 is  
significantly less than the amount used in version 1.2.x.  
Average block size RAM usage for each 80 RAM usage for each 100 Minimum system RAM  
(bytes)  
GB IO Accelerator  
(Megabytes)  
GB IO Accelerator  
(Megabytes)  
requirement for 320 GB  
MezzIO Accelerator*  
250  
280  
1 GB  
8,192  
4,096  
2,048  
1,024  
512  
400  
530  
1.6 GB  
3 GB  
750  
1,030  
2,000  
3,970  
1,450  
2,850  
5.8 GB  
11.4 GB  
Average block size Minimum system RAM  
Minimum system RAM  
Minimum system RAM  
(bytes)  
requirement for 640 GB requirement for 785 GB requirement for 1.2 TB  
MezzIO Accelerator*  
MezzIO Accelerator*  
MezzIO Accelerator*  
2 GB  
2.2 GB  
3.4 GB  
8,192  
4,096  
2,048  
1,024  
512  
3.2 GB  
4.2 GB  
6.4 GB  
6 GB  
8.1 GB  
12.4 GB  
11.6 GB  
15.7 GB  
24 GB  
22.8 GB  
31.2 GB  
47.6 GB  
* For IO Accelerator use only. Additional RAM is needed for system operating system and applications.  
HP PCIe IO Accelerator Duo requirements  
In addition to the IO Accelerator cooling and RAM requirements listed in the previous table, the IO  
Accelerator Duo requires at least:  
A PCIe Gen1 x8 slot or a PCIe Gen2 x4 slot  
A minimum of a full-height, half-length slot with a x8 physical connection. For systems with PCI 1.1, all  
eight signaling lanes must be active for full IO Accelerator Duo performance. For systems with PCIe 2.0,  
only four signaling lanes must be active for full performance.  
NOTE: With driver version 3.1.0 and later, the driver detects in the BIOS if the PCIe slot supports  
a 75W power draw. If the slot supports up to 75W, the IO Accelerator device draws up to that  
amount of power. However, if an external power cable is used, power is only supplied by that  
cable.  
To verify whether a slot is supplying 75W, view the system logs or use the fio-pci-check  
utility.  
Introduction 11  
Download from Www.Somanuals.com. All Manuals Search And Download.  
Windows Server environments  
Installation overview  
CAUTION: The firmware for all IO Accelerator devices in a system running HP IO Accelerator  
VSL 3.1.x or later must be upgraded to the latest version of the firmware.  
CAUTION: If you have IO Accelerator devices configured for HP IO Accelerator VSL 2.x or  
earlier, you must upgrade the firmware before installing new devices in the system. For upgrade  
procedures, see "Upgrading device firmware from VSL 1.x.x or 2.x.x to 3.x.x (on page 19)."  
1.  
If you are installing this version of HP IO Accelerator VSL on a system with IO Accelerator devices  
configured for VSL 2.x or earlier, follow the instructions in "Upgrading device firmware from VSL 1.x.x  
or 2.x.x to 3.x.x (on page 19)."  
2.  
3.  
4.  
Install ("Installing software on a Windows operating system" on page 13) the latest version of the IO  
Accelerator driver.  
If necessary, upgrade the firmware ("Upgrading the device firmware using Windows" on page 23) to  
the latest version.  
Configure the devices by adding a file system ("Adding a file system, formatting, and performing  
multi-disk configuration" on page 28), creating a RAID configuration (on page 29), and so on.  
Uninstalling a previously-installed driver  
CAUTION: Upgrading an IO Accelerator from Version 1.2.x to Version 2.x is a one-way  
process. New firmware is required that is not compatible with 1.2.x systems. Be sure all data is  
backed up before continuing with the upgrade.  
Before installing a new driver, uninstall the previous version.  
1.  
2.  
3.  
4.  
Go to the Start menu.  
Open the Control Panel.  
Select Add or Remove Programs.  
Select the package or program.  
NOTE: Uninstall IO Manager and HP StorageWorks IO Accelerator software.  
5.  
6.  
Select Remove.  
Reboot the server, and then log in as a user with Administrative privileges.  
If previous (downrev) drivers (such as driver files fiodrive.sys) exist on the system (typically located in  
C:\Windows\System32\drivers) after you have uninstalled the software, then upon reboot, Windows  
Windows Server environments 12  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
recognizes those drivers for the IO Accelerator. This situation causes issues with the installation of Version  
3.x.  
After the software is uninstalled and the system is rebooted, Windows operating systems display a New  
Hardware Foundmessage and normally do not find a driver for the new device.  
If a driver is found, go to Control Panel > Device Manager and search for HP IO Accelerator. View the driver  
details and check the revision of the driver. If the HP IO Accelerator appears in the Device Manager and the  
driver revision is not 3.x or later, manually uninstall the driver for each IO Accelerator in the system, and then  
reboot the system.  
If no driver is available, the HP IO Accelerator appears in the Device Manager as a mass storage controller.  
The system must be in this state before you install the 3.x software.  
Installing software on a Windows operating system  
1.  
2.  
3.  
Complete the hardware installation.  
Log in to your computer as Administrator or have Administrator rights.  
Download the Windows® IO Accelerator installation program from the HP website  
(http://www.hp.com/support) to your desktop or a convenient directory.  
4.  
Run the IO Accelerator installation program (Setup Wizard) as explained in "Using the Setup Wizard."  
CAUTION: The installation program does not remove previous versions of the software. If a  
previous version is removed after the newest version is installed, the driver does not load after a  
restart. If this occurs, run the Repair option in the installation program from Programs and  
Features (or Add or Remove Programs) in the Control Panel, and then restart the computer.  
NOTE: Screens might appear slightly different (for example, the word "StorageWorks" might not  
appear in the name or pathname with version 2.x).  
Windows Server environments 13  
Download from Www.Somanuals.com. All Manuals Search And Download.  
   
Using the Setup Wizard  
1.  
Download the IO Accelerator installation program. The Welcome screen appears.  
2.  
Click Next. The Installation Recommendations screen appears.  
Windows Server environments 14  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
3.  
If you have a previous version of the IO Accelerator software installed, follow the directions on the  
screen to uninstall it. Otherwise, click Next. The Select Installation Folder screen appears.  
4.  
If you want to use the default installation location (C:\Program Files\HP IO Accelerator), click Next. If  
you want to change the default location, click Browse, navigate to the folder you want, and then click  
Next. The Custom Setup screen appears.  
5.  
If you want to use the default program features for the installation, click Next. If you want to customize  
the features, click the plus sign next to Driver and Utilities or IO Management Utilities, and then select the  
customizations you want from the list.  
Windows Server environments 15  
Download from Www.Somanuals.com. All Manuals Search And Download.  
CAUTION: Changing the default features might affect the performance or functionality of the IO  
Accelerator software.  
The following features are available:  
o
o
o
o
Device Driver (required for the drive to be recognized in Windows® operating system)  
HP IO Accelerator Management Tool (Java® Runtime Software)  
SNMP (requires Microsoft® SNMP to be installed)  
Utilities (Command Line Utilities)  
HP recommends that you select will be installed on local hard drive for each feature. If you change your  
mind later, you can use the Repair option in Programs and Features, or Add or Remove Programs in the  
Control Panel.  
Windows Server environments 16  
Download from Www.Somanuals.com. All Manuals Search And Download.  
The Configure Shortcuts screen appears.  
6.  
If you want to accept the default program shortcuts, click Next. Otherwise, clear any options you do not  
want, and then click Next. The Ready to Install screen appears.  
7.  
Click Install to begin installation. An installation status screen appears. When installation is complete,  
the Completed screen appears.  
NOTE: You might be prompted to stop installation and start the Microsoft® SNMP service during  
the software installation.  
Windows Server environments 17  
Download from Www.Somanuals.com. All Manuals Search And Download.  
8.  
Click Finish to exit the Wizard and launch the IO Accelerator software.  
The installation program performs the following:  
Creates a folder for the IO Accelerator software components (the default path is C:\Program Files\HP  
IO Accelerator).  
Windows Server environments 18  
Download from Www.Somanuals.com. All Manuals Search And Download.  
Installs and loads the IO Accelerator Windows® driver. (When complete, you will need to reboot the  
system.)  
Installs support for SNMP (if installed, and if Microsoft® SNMP is installed and the SNMP service is  
running).  
When the installation program creates the IO Accelerator folder on the drive, it also creates these sub-folders:  
<driver version>\Driver—Installs manually using Device Manager.  
Firmware—Contains the latest IO Accelerator firmware.  
SMI-S—Currently not used; reserved for future SMI-S use.  
SNMP—Contains the SNMP components.  
Utils—Contains the IO Accelerator command line utilities.  
To continue installation, go to "Upgrading the device firmware using Windows (on page 23)."  
Upgrading device firmware from VSL 1.x.x or 2.x.x to  
3.x.x  
CAUTION: You cannot downgrade an HP IO Accelerator device firmware to an earlier version  
after you have upgraded the device.  
CAUTION: Upgrading IO Accelerator devices that were previously configured for VSL 1.x.x or  
2.x.x to work with VSL 3.x.x requires a low-level media format of the device. No user data is  
maintained during the media format process. Be sure to backup all data on your IO Accelerator  
device as instructed before upgrading the firmware.  
Version 3.1.x of the HP IO Accelerator VSL supports new features, including the latest generation of IO  
Accelerator architecture and improved Flashback protection. These features require the latest version of the  
firmware. Every IO Accelerator device in a system running 3.1.x or later must be upgraded to the latest  
version of the firmware.  
For example, if you have a system running 2.3.1 HP IO Accelerator VSL with IO Accelerator devices  
previously installed, and you want to install new IO Accelerator Gen2 devices (that require the latest version  
of the firmware), then you will need to upgrade all of the existing devices to the latest firmware version.  
Upgrade path  
Depending on the current version of your HP IO Accelerator device, to preserve the internal structure of the  
device, you might have to perform multiple upgrades. The following path is the minimum upgrade path that  
you must follow. Upgrade the HP IO Accelerator VSL software on the system, and upgrade the firmware to  
the compatible version in the following order:  
1.2.4 > 1.2.7 > 2.1.0 > 2.2.3 > 3.1.x  
For VSL upgrade information for the HP IO Accelerator, see the HP IO Accelerator Release Notes on the HP  
website (http://www8.hp.com/us/en/support-drivers.html). General upgrade instructions, including the  
firmware update instructions, are available in the HP IO Accelerator User Guide for each operating system.  
Overformatting not supported  
Windows Server environments 19  
Download from Www.Somanuals.com. All Manuals Search And Download.  
   
The –o overformatoption is not supported in the 3.x.x VSL software. All upgraded HP IO Accelerator  
devices are formatted to the maximum advertised capacity, regardless of whether the device was  
overformatted prior to the upgrade.  
Upgrading procedure  
Be sure to follow the upgrade path and make sure that all previously installed IO Accelerator devices are  
updated with the appropriate 2.3.1-compatible firmware.  
1.  
Prepare each existing IO Accelerator device for upgrade.  
a. Backup user data on each IO Accelerator device.  
CAUTION: Upgrading IO Accelerator devices that were previously configured for VSL 1.x.x or  
2.x.x to work with VSL 3.x.x requires a low-level media format of the device. No user data is  
maintained during the media format process. Be sure to backup all data on your IO Accelerator  
device as instructed before upgrading the firmware.  
Do not back up the data onto another IO Accelerator device on the same system. The backup must  
be to a local disk or to an externally attached volume.  
b. Run the fio-bugreportcommand-line utility and save the output. The output captures the device  
information for each device in the system. This device information will be useful in troubleshooting  
any upgrade issues. For example:  
fio-bugreport  
c. Detach the ioDrive devices. For example:  
fio-detach /dev/fct*  
For more information, see fio-detach(on page 45).  
2.  
3.  
Unload the current IO Accelerator driver. For example:  
$ modprobe -r iomemory-vsl  
Uninstall the 2.x HP IO Accelerator VSL software.  
a. Go to Start > Control Panel.  
b. Click Administrative Tools.  
c. Click Computer Management.  
d. Click Device Manager in the console tree at the left.  
e. Expand the IO Accelerator Devices item.  
f. Right-click the desired IO Accelerator device.  
g. Click Uninstall or Delete the driver software for this device (depending on your version of  
Windows®).  
h. Restart the computer.  
4.  
Install the new VSL.  
a. Download the IO Accelerator installation program for Microsoft Windows from the HP website  
(http://www.hp.com/support) to your desktop or a convenient directory.  
b. Run the IO Accelerator installation program. The installation program presents a custom setup  
tree-view with options for installation.  
Mouse over a component in the tree view to see the description.  
c. Click Next.  
Windows Server environments 20  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
d. To select a different folder for the installation, browse to the folder and click OK. The default folder  
is C:\Program Files\HP IO Accelerator.  
The uninstaller file is placed in the root of the HP IO Accelerator (default install folder).  
e. Follow the onscreen prompts to complete the install.  
f. Choose Reboot Now on the finish screen of the installer.  
For full installation instructions, see "Installing software on a Windows system ("Installing software  
5.  
Update the firmware on each device to the latest version using fio-update-iodrive.  
CAUTION: Do not turn off the power during a firmware upgrade, because this might cause  
device failure. If a UPS is not in place, consider adding one to the system before performing a  
firmware upgrade.  
For all warnings, alerts, and options pertaining to this command utility, see "fio-update-iodrive."  
Sample syntax:  
fio-update-iodrive <iodrive_version.fff>  
Where <iodrive_version.fff>is the path to the firmware archive. This command updates all of  
the devices to the selected firmware. To update specific devices, consult the utility reference for more  
options.  
6.  
7.  
Reboot the system  
If you run the fio-statuscommand, a warning appears, indicating that the upgraded devices are  
missing a lebmap. This is a customary warning, and the issue will be corrected in the next step.  
Format each device using fio-format. For example:  
fio-format <device>  
CAUTION: Use this utility with care since it deletes all user information on the IO Accelerator.  
You are prompted to confirm that you want to erase all data on the device. The format might take an  
extended period of time, depending on the wear on the device.  
8.  
9.  
Attach all IO Accelerator devices. For example:  
fio-attach /dev/fct*  
Verify the status of all devices using fio-status. For example:  
fio-status -a  
Your IO Accelerator devices are now upgraded for this version of the HP IO Accelerator. You can now install  
any IO Accelerator Gen2 devices.  
Manual installation on Windows Server  
The Windows® Setup program installs your driver and software. However, if the driver does not install, you  
can perform a manual installation (or upgrade) using the following steps.  
Windows Server environments 21  
Download from Www.Somanuals.com. All Manuals Search And Download.  
   
Manually installing on Windows Server 2003  
The Windows® Driver Wizard automatically detects the new IO Accelerator and begins to locate its driver  
after you restart the system. First, Windows® prompts you to locate the software driver.  
1.  
If you have not done so, download and run the IO Accelerator Windows® Setup program from the HP  
2.  
3.  
4.  
5.  
6.  
Return to the Update Driver dialog box.  
Select Install from a list or specific location. Click Next.  
Select Don't Search. I will choose the driver to install. Click Next.  
To open a browsing dialog, click Have Disk.  
Browse to the folder that contains the IO Accelerator driver. The default is C:\ProgramFiles\HPIO  
Accelerator\Driver.  
7.  
8.  
Click OK. The correct driver installs the device software. When the driver installation completes, the  
following message appears:  
The wizard has finished installing the software for: ioMemory VSL Device  
To continue installation, see "Upgrading the device firmware using Windows (on page 23)."  
Windows Server environments 22  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Manually installing on Windows Server 2008 and 2012  
The Windows Driver Wizard automatically detects the new IO Accelerator and begins to locate its driver  
after you restart the system. First, the OS prompts you to locate the software driver.  
1.  
2.  
Click Browse next to the path field. A file dialog box appears.  
Select the folder that contains the IO Accelerator driver. The default is C:\Program Files\HP  
StorageWorks IO Accelerator\Drivers.  
3.  
4.  
5.  
6.  
Click OK.  
Click Next. The correct driver installs the device software.  
When the driver installation completes, restart the computer.  
To continue installation, see "Upgrading the device firmware using Windows (on page 23)."  
Upgrading the device firmware using Windows  
CAUTION: Upgrade the device firmware only if the System Event Log reports out-of-date  
firmware, or if you are instructed to do so by HP Customer Support or the Release Notes  
document.  
Viewing the firmware version  
The firmware version can be found in the Windows® System Event Log. The version number is reported by  
the IO Accelerator Management Tool and the fio-statuscommand line utility. For more information, see  
the HP IO Accelerator Management Tool User Guide or "fio-status (on page 47)."  
Performing the upgrade  
Windows Server environments 23  
Download from Www.Somanuals.com. All Manuals Search And Download.  
   
CAUTION: Do not turn off the power during a firmware upgrade, because this might cause  
device failure. If a UPS is not in place, consider adding one to the system before performing a  
firmware upgrade.  
NOTE: You might be required to upgrade the IO Accelerator driver if you upgrade the device  
firmware. For more information, see the Release Notes.  
Upgrading the firmware might take several minutes. The update command displays a progress  
bar to indicate the pace of the upgrade. This progress bar might appear to stall at points, but let  
the firmware update command complete.  
Typically, you use the HP IO Accelerator Management Tool console to perform firmware upgrades. You can  
also perform the upgrade using the fio-update-iodrivecommand line utility.  
To upgrade the firmware on the IO Accelerator using fio-update-iodrive:  
1.  
2.  
Back up the data on the IO Accelerator prior to any upgrade as a precaution.  
Open a command prompt and navigate to the C:\Program Files\HP IO Accelerator\Utils  
directory.  
3.  
Run the Detachcommand for your device at the command prompt.  
For example:  
fio-detach /dev/fct0  
detaches the IO Accelerator labeled fct0.  
a. Specify the name of the drive you want to detach. Each IO Accelerator is labeled /dev/fct#  
where # is its bus number.  
b. Use the fio-statuscommand at a command prompt to view this number.  
The fio-detachcommand shows a progress bar and percentage as it completes the operation.  
4.  
Type the following command at the command prompt:  
fio-update-iodrive <path> ioaccelerator_<version>.fff -d /dev/fct#  
where <path>is the location of the IO Accelerator firmware (the default is C:\Program Files\HP  
IOAccelerator\Firmware) and <version> is the numerical version of the driver (higher is newer.)  
The -d/dev/fct#option identifies the IO Accelerator to update, where # is its bus number. Use the  
fio-statuscommand at a command prompt to view this number.  
All three external LED indicators light up during the update process.  
Restart the computer to finish the firmware upgrade.  
5.  
Upgrading driver software using Windows  
Upgrading the driver software in Windows® operating systems depends on whether you are using a  
non-RAID or a RAID configuration.  
Upgrading driver software using Windows in a non-RAID  
configuration  
1.  
See the Release Notes for your version of the driver for details on any additional steps to perform the  
upgrade.  
Windows Server environments 24  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
CAUTION: Be sure to read the Release Notes document that comes with each new release as  
well as these installation instructions to ensure no loss of data when performing upgrades.  
2.  
3.  
4.  
5.  
6.  
7.  
8.  
9.  
Follow the steps in "Uninstalling the IO Accelerator driver" uninstall the existing driver.  
Download the latest driver from the HP website (http://www.hp.com/support).  
Either unzip or run the Windows® package to copy the files to a convenient directory.  
Go to Start > Control Panel.  
Click Administrative Tools.  
Click Computer Management.  
In the console tree at the left, click Device Manager.  
Expand HP IO Accelerators.  
10. Select System Devices with pre-2.1 drivers.  
11. Right-click an IO Accelerator.  
12. Click Update Driver Software. If necessary, see "Manual installation on Windows Server (on page 21)"  
for details on the remaining steps to install the updated driver. The Windows® operating system now  
detects your IO Accelerator.  
Upgrading driver software using Windows in a RAID  
configuration  
1.  
See the Release Notes for your version of the driver for details on any additional steps to perform the  
upgrade.  
CAUTION: Be sure to read the Release Notes document that comes with each new release as  
well as these installation instructions to ensure no loss of data when performing upgrades.  
2.  
3.  
Shut down any applications that are accessing the IO Accelerator devices.  
Open the IO Accelerator Utilities folder. (The default location for this release is C:\Program  
Files\HP IO Accelerator\Utils.)  
4.  
Double-click the AutoAttachDisable.reg file to add a key to the Windows® registry. This key prevents  
the IO Accelerator device from automatically attaching the next time you restart the computer.  
5.  
6.  
7.  
8.  
9.  
Uninstall the IO Accelerator software in Windows® Add/Remove Programs.  
Restart the computer.  
Download the latest driver from the HP website (http://www.hp.com/support).  
Either unzip or run the Windows® package to copy the files to a convenient directory.  
Open the IO Accelerator Utilities folder. (The default location is C:\Program Files\HP IO  
Accelerator\Utils.)  
10. To reset the key in the Windows® registry, double-click the AutoAttachEnable.reg file. Your drive will  
automatically attach the next time you restart the computer.  
11. Update the firmware of the devices. Follow the steps in "Upgrading the device firmware using  
Windows (on page 23)."  
Restart the computer after the firmware upgrade is complete. The IO Accelerator Check Utility will run at next  
boot.  
Windows Server environments 25  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Your Windows® operating system now detects the devices in the RAID configuration with the upgraded  
software.  
Defragmentation  
The IO Accelerator does not need to be defragmented. However, some Windows® versions automatically  
run defragmentation as a scheduled task. If necessary, turn off automatic defragmentation.  
Silent install option  
Uninstalling the previous version  
If you have a version of the HP IO Accelerator previously installed, you must uninstall it first. You must  
manually reboot the computer after installing the new version with the silent install option. This step must be  
performed prior to using any HP IO Accelerator utilities or functionality.  
If you are installing remotely or with scripts, you can use the silent install option (/quiet) when you run the  
installation program in the command-line interface.  
In the command-line interface, navigate to the folder that contains the .exeinstaller file, and then run the  
following command:  
<installname>.exe /quiet  
Where <installname>.exeis the name of the installer file.  
This option installs the HP IO Accelerator using the default settings and eliminates the need to click Next or  
select settings during installation.  
Installing components  
The default command-line installation installs the HP IO Accelerator (including the command-line utilities) and  
the firmware file. You can add additional components using the following command:  
<installname>.exe /quiet ADDLOCAL=SNMP,SDK,SMIS  
Remove any component (and leading or trailing comma) that you do not want to install.  
Performing a silent uninstall  
You can silently uninstall the HP IO Accelerator with this command:  
<installname>.exe /uninstall /quiet  
Outdated firmware check  
Checking by using the command-line interface  
1.  
Run fio-statusand examine the output:  
o
o
If any device is in minimal mode, then the firmware is outdated.  
If the firmware listed for any device is a lower number than the latest firmware version as noted in  
the release notes, then the firmware is old, but not outdated.  
2.  
If the firmware is old or outdated, update it using the fio-update-iodriveutility.  
Checking by using the optional GUI interface  
Windows Server environments 26  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
You can use the (optional) HP IO Accelerator Management Tool GUI program to check for outdated  
firmware.  
To check for outdated or old firmware:  
1.  
Launch the IO Accelerator Management Tool application and look for any devices that have a warning  
symbol.  
2.  
3.  
To ensure that the alert is from outdated firmware, click any devices with a warning symbol.  
Select all devices requiring firmware update and use the IO Accelerator Management Tool to update  
the firmware. For details, see the HP IO Accelerator Management Tool User Guide.  
To check for old but not outdated firmware:  
1.  
2.  
Find the name of the latest firmware version as noted in the release notes.  
Use the IO Accelerator Management Tool application to check the firmware version of each IO  
Accelerator device against the latest firmware version.  
For instructions on how to update the firmware, see the HP IO Accelerator Management Tool User  
Guide.  
Enabling PCIe power  
For PCIe IO Accelerators, if you have installed any dual IO Accelerator devices, such as the HP ioDrive2  
Duo, then the device might require additional power than the minimum 25 W provided by PCIe Gen2 slots  
to properly function.  
For instructions on enabling the device to draw additional power from the PCIe slots, see "Enabling PCIe  
power override (on page 38)."  
IO Accelerator naming  
The IO Accelerator receives an identification name and number as part of the installation process. The syntax  
is fctx, where xis the number of the bus where you installed the IO Accelerator. Use the IO Accelerator  
Management Tool to view this bus number, or follow these steps:  
1.  
Select Start > Control Panel > System > Hardware > Device Manager.  
Windows Server environments 27  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
2.  
Expand HP StorageWorks IO Accelerators, and then double-click HP StorageWorks IO Accelerator. The  
Properties dialog box appears. The Location field shows the bus number for your device (fct8in the  
following example).  
NOTE: HP assigns bus numbers, which begin at 0. These numbers might or might not reflect the  
physical location of the bus. For example, the second slot from the edge of the motherboard may  
be Bus 2, but it might also be Bus 16, or another arbitrary number. Checking Device Manager is  
one way to confirm the specific bus number for your installation. You can also use IO Accelerator  
Management Tool to view this number.  
Adding a file system, formatting, and performing  
multi-disk configuration  
If you want to use a standard file system such as NTFS, you can create and format partitions using standard  
disk management utilities.  
For example, to initialize the IO Accelerator:  
1.  
From the Start Menu, select Control Panel > Administrative Tools > Computer Management, and then  
from the Storage branch of the tree, select Disk Management.  
2.  
Find the IO Accelerator in the list of storage devices, right-click it, and then select Initialize Disk.  
Windows Server environments 28  
Download from Www.Somanuals.com. All Manuals Search And Download.  
   
3.  
After the device is initialized, you can add partitions or volumes. For more information, see the Disk  
Management documentation.  
You can create dynamic volumes, which can be used in multi-disk RAID configurations (spanned, striped, or  
mirrored). For more information, see the Disk Management documentation.  
Creating a RAID configuration  
You can use your IO Accelerator as part of a RAID configuration. To do so, you must format the IO  
Accelerator as a dynamic volume. You can then use this dynamic volume to create multi-disk RAID  
configurations (spanned, striped, or mirrored).  
For specific steps to perform a RAID configuration, see the Windows® Disk Management Utility  
documentation.  
NOTE: If you are using RAID1 mirroring and one device fails, enter the fio-formatcommand  
on the replacement device (not the existing good device) before rebuilding the RAID.  
HP recommends that you do not use a RAID5 configuration.  
Using the IO Accelerator as swap with Windows  
operating system  
To safely use the IO Accelerator as swap space, you must use the fio-configutility to pass a special  
pre-allocation parameter.  
For example:  
fio-config -p FIO_PREALLOCATE_MEMORY 1072,4997,6710,10345  
where 1072,4997,6710,10345are serial numbers obtained from fio-status.Be sure to use serial  
numbers for the IO Accelerator modules and not the adapter.  
A 4K sector size format is required for swap. This reduces the driver memory footprint.  
CAUTION: You must have 400MB of free RAM per 80GB of IO Accelerator capacity (formatted  
to 4KB block size) to enable the IO Accelerator with pre-allocation enabled for swap. Attaching  
an IO Accelerator with pre-allocation enabled and insufficient RAM might result in the loss of user  
processes and system instability.  
Setting up SNMP for Windows operating systems  
Use the following procedures to set up SNMP for Windows® operating systems.  
SNMP details for Windows operating systems  
You can set up a test mode with your IO Accelerator Windows® SNMP agent. Doing so enables you to set  
test values in a Windows® registry and force SNMP traps without having to create the actual conditions on  
the device.  
For example, you can use the SNMP test mode to change the non-writeable indicator and generate a trap,  
or simulate a change to the physical or logical size of the device, and so forth.  
Windows Server environments 29  
Download from Www.Somanuals.com. All Manuals Search And Download.  
     
NOTE: To use SNMP test mode, you must have installed the SNMP option with your PCIe IO  
Accelerator.  
Using test mode registry values  
The following table shows the registry entries included for SNMP test values. Entries marked by an asterisk (*)  
generate SNMP traps when set to the indicated values, and the  
cpqDrvMIBCondition and cpqDrvInfoStatus MIBvariables might be affected because of the  
changes.  
NOTE: All entries, except those marked by **, reflect your registry changes immediately. Entries  
marked by ** require a restart of the Windows® SNMP agent for the changes to take effect.  
SNMP test registry entry  
Description  
Lower word of the available logical capacity in bytes  
cpqioDrvExtnAvailLogCapacityL  
Upper word of the available logical capacity in bytes  
Lower word of the total number of bytes read since the device was formatted  
Upper word of the total number of bytes read since the device was formatted  
Lower word of the number of user data bytes written  
cpqioDrvExtnAvailLogCapacityU  
cpqioDrvExtnBytesReadL  
cpqioDrvExtnBytesReadU  
cpqioDrvExtnLogBytesWrittenL  
cpqioDrvExtnLogBytesWrittenU  
cpqioDrvExtnPhysBytesWrittenL  
cpqioDrvExtnPhysBytesWrittenU  
cpqioDrvExtnTotalLogCapacityL  
cpqioDrvExtnTotalLogCapacityU  
cpqioDrvExtnTotalPhysCapacityL  
cpqioDrvExtnTotalPhysCapacityU  
Upper word of the number of user data bytes written  
Lower word of the total physical bytes written  
Upper word of the total physical bytes written  
Lower word of the total logical capacity in bytes as formatted  
Upper word of the total logical capacity in bytes as formatted  
Lower word of the total logical capacity in bytes as formatted  
Upper word of the total logical capacity in bytes as formatted  
Lower word of the useable physical capacity in bytes. This space is holding  
valid data, or is erased and ready for writing, or is waiting to be reclaimed  
via garbage collection.  
cpqioDrvExtnUsablePhysCapacity  
L
Upper word of the useable physical capacity in bytes. This space is holding  
valid data, or is erased and ready for writing, or is waiting to be reclaimed  
via garbage collection.  
cpqioDrvExtnUsablePhysCapacity  
U
Current temperature of the device in degrees Celsius. If this value is set  
above 78°C for ioDimm cards, a trap is generated. If set above 90°C for HP  
Mezzanine cards, a trap is generated.  
cpqioDrvInfoInternalTemp*  
1 = flashback redundancy is degraded; 2 = false  
cpqioDrvInfoFlashbackIndicator*  
1 = device is no longer writable because it has surpassed the read-only  
threshold; 2 = false  
cpqioDrvInfoNonWritableIndicato  
r*  
Upper word of the total logical capacity in bytes as formatted  
Current state of the attached client device:  
cpqioDrvInfoPercentLifeRemaining  
cpqioDrvInfoState*  
(trap generated if state = 4)  
unknown(0)  
detached(1)  
attached(2)  
minimal(3)  
error(4)  
detaching(5)  
attaching(6)  
scanning(7)  
Windows Server environments 30  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
SNMP test registry entry  
Description  
formatting(8)  
updating(9)  
attach(10)  
detach(11)  
format(12)  
update(13)  
Boolean: True = device has surpassed the wearout threshold.  
Set test mode on or off  
cpqioDrvInfoWearoutIndicator*  
cpqioDrvTestMode  
cpqioDrvTestModeIndex  
mib  
Number indicating the selected IO Accelerator  
Name of the MIB in use  
Path to the driver, set at installation  
Set trap generation on or off  
pathname  
traps**  
Number of milliseconds to wait until getting the next value from the driver to  
generate a trap  
updateDelay**  
SNMP MIB fields supporting Windows Server  
cpqIoDimmMibRevMajor  
cpqioDimmMibRevMinor  
cpqIoDimmMibCondition  
cpqIoDimmInfoIndex  
cpqIoDimmInfoStatus  
cpqIoDimmInfoName  
cpqIoDimmInfoSerialNumber  
cpqIoDimmInfoPartNumber  
cpqIoDimmInfoSubvendorPartNumber  
cpqIoDimmInfoSparePartNumber  
cpqIoDimmInfoAssemblyNumber  
cpqIoDimmInfoFirmwareVersion  
cpqIoDimmInfoDriverVersion  
cpqIoDimmInfoUID  
cpqIoDimmInfoState  
cpqIoDimmInfoClientDeviceName  
cpqIoDimmInfoBeacon  
cpqIoDimmInfoPCIAddress  
cpqIoDimmInfoPCIDeviceID  
cpqIoDimmInfoPCISubdeviceID  
cpqIoDimmInfoPCIVendorID  
cpqIoDimmInfoPCISubvendorID  
Windows Server environments 31  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
cpqIoDimmInfoPCISlot  
cpqIoDimmInfoWearoutIndicator  
cpqIoDimmInfoFlashbackIndicator  
cpqIoDimmInfoWritableIndicator  
cpqIoDimmInfoInternalTemp  
cpqIoDimmInfoHealthPercentage  
cpqIoDimmInfoMinimalModeReason  
cpqIoDimmInfoReducedWriteReason  
cpqIoDimmInfoMilliVolts  
cpqIoDimmInfoMilliVoltsPeak  
cpqIoDimmInfoMilliVoltsMin  
cpqIoDimmInfoMilliWatts  
cpqIoDimmInfoMilliWattsPeak  
cpqIoDimmInfoMilliAmps  
cpqIoDimmInfoMilliAmpsPeak  
cpqIoDimmInfoAdapterType  
cpqIoDimmInfoAdapterPort  
cpqIoDimmInfoAdapterSerialNumber  
cpqIoDimmInfoAdapterExtPowerPresent  
cpqIoDimmInfoPowerlossProtectDisabled  
cpqIoDimmInfoInternalTempHigh  
cpqIoDimmInfoAmbientTemp  
cpqIoDimmInfoPCIBandwidthCompatibility  
cpqIoDimmInfoPCIPowerCompatibility  
cpqIoDimmInfoActualGoverningLevel  
cpqIoDimmInfoLifespanGoverningLevel  
cpqIoDimmInfoPowerGoverningLevel  
cpqIoDimmInfoThermalGoverningLevel  
cpqIoDimmInfoLifespanGoverningEnabled  
cpqIoDimmInfoLifespanGoverningTgtDate  
cpqIoDimmExtnIndex  
cpqIoDimmExtnTotalPhyslCapacityU  
cpqIoDimmExtnTotalPhysCapacityL  
cpqIoDimmExtnTotalLogCapacityU  
cpqIoDimmExtnTotalLogCapacityL  
cpqIoDimmExtnBytesReadU  
Windows Server environments 32  
Download from Www.Somanuals.com. All Manuals Search And Download.  
cpqIoDimmExtnBytesReadL  
cpqIoDimmExtnBytesWrittenU  
cpqIoDimmExtnBytesWrittenL  
cpqIoDimmExtnFormattedBlockSize  
cpqIoDimmExtnCurrentRAMUsageU  
cpqIoDimmExtnCurrentRAMUsageL  
cpqIoDimmExtnPeakRAMUsageU  
cpqIoDimmExtnPeakRAMUsageL  
cpqIoDimmWearoutTrap  
cpqIoDimmNonWritableTrap  
cpqIoDimmFlashbackTrap  
cpqIoDimmTempHighTrap  
cpqIoDimmTempOkTrap  
cpqIoDimmErrorTrap  
cpqIoDimmPowerlossProtectTrap  
Using installation logs  
You can create a Windows® installation log to track various types of messages displayed when installing the  
driver.  
Windows Installer logging options  
The Windows® Installer handles its installations through Msiexec.exe. The logging command is issued with  
the /Lparameter. Options are indicated with an asterisk. For example, the following command turns on  
logging with verbose output (*v):  
msiexec /i "C:\MyPackage\Example.msi" /L*V "C:\log\example.log"  
The following logging options enable you to create different types of logs, depending on the information you  
need about the installation.  
Option  
Description  
Startup of actions  
*a  
*c  
*e  
*i  
Initial UI parameters  
All error messages  
Status messages  
Out of memory or fatal exit information  
Out-of-disk-space messages  
Terminal properties  
Action-specific records  
User requests  
*m  
*o  
*p  
*r  
*u  
*v  
*w  
Verbose output  
Nonfatal warnings  
Windows Server environments 33  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Option  
Description  
Extra debugging information  
Append to existing log file  
Flush each line to the log  
*x  
*+  
*!  
Log all information except for v and x options  
**  
Creating an installation log  
1.  
2.  
3.  
4.  
Determine the path of the MSI file (for example, C:\MyPackage\Example.exe).  
Decide the path of the log (for example, C:\log\example.log).  
Open cmd.exe.You can use any command shell.  
Use the msiexeccommand line to launch the MSI with logging parameters.  
Any installation logging command line must have this form:  
msiexec /i <path_to_msi> /L*V <path_to_log>  
The following is a sample command for creating the installation log. The /iparameter launches the MSI  
package. Be sure to specify the actual paths to your MSI file and log file.  
msiexec /i "C:\MyPackage\Example.msi" /L*V "C:\log\example.log"  
If you want the log to be created next to the MSI, you can specify only the name of the log file:  
msiexec /i "C:\MyPackage\Example.msi" /L*V "example.log"  
After the installation is finished, the log is complete.  
Using an .exe bootstrapper  
When the installation package is included in an .exe bootstrapper, do not use msiexec. in the command  
line. For example:  
"C:\MyPackage\Setup.msi" /L*V "example.log"  
Creating an uninstall log  
Launch the installer with the logging command, for example:  
"C:\MyPackage\Setup.exe" /L*V "example.log"  
This command puts the package into maintenance mode, and you can choose to uninstall it. Because the  
package was launched with logging, an uninstall log is generated.  
Creating a patch install log  
To create a log for a patch installation, follow the procedure in "Creating an installation log (on page 34),"  
but replace the /iparameter with /p,for example:  
msiexec /p "C:\MyPackage\Patch.msp" /L*V "C:\log\patch.log"  
Using the .exe boostrapper  
Another approach is to create a log file by using the /L*Vparameters in the command line of the Installer  
Bootstrapper. Also, these parameters can be always passed to the MSI when the package is launched  
through the .exe bootstrapper.  
If you want your installation package to create a log, follow these steps:  
Windows Server environments 34  
Download from Www.Somanuals.com. All Manuals Search And Download.  
   
1.  
2.  
3.  
4.  
Open the Installer project of your installation package.  
Go to the Media page and select the Bootstrapper tab.  
Select the Create EXE setup file option.  
Set the MSI Command Line field to /L*V "C:\package.log".  
When the user launches the installation through the bootstrapper, a log of the installation ("package.log")  
will be created automatically in the "C:\"drive.  
The command line received by the bootstrapper overrides the command line in the MSI Command Line field.  
Therefore, if you launch an EXE package with logging parameters, these parameters will be used for creating  
the log.  
Automated logging with the Windows Installer Logging Policy  
To enable automated logging for debugging purposes, create a registry value.  
1.  
2.  
From a command line, run regedit.exe.  
Locate the following registry key:  
HKEY_LOCAL_MACHINE\Software\Policies\Microsoft\Windows\Installer  
3.  
4.  
Create a new STRING Entry named Logging.  
Change its value using the logging options discussed previously. Note that you cannot use the wildcard  
()for this approach.  
The .LOGfile will be created in the Tempfolder of the current user account with a name of this form:  
MSI.LOG.  
NOTE: Do not leave this option active because every install or uninstall operation on an .msi  
file will create a new log. This action results in unnecessarily occupied disk space and makes your  
system crowded. Only use this option for debugging purposes.  
The logging policy is particularly useful for troubleshooting Active Directory and Group Policy deployments.  
In these cases, the installation is carried out without a user interface, and you cannot specify a command line  
for the .msi.  
Troubleshooting event log messages  
The Windows® System Event Log displays fiodrivemessages concerning the IO Accelerator. These  
messages can be either informational and warnings.  
NOTE: Each IO Accelerator is numbered from 0 upwards. These numbers reflect the bus number  
where you installed the device. Use the IO Accelerator Management Tool to view this number for  
your device.  
To open the Windows Event Viewer:  
1.  
2.  
3.  
4.  
5.  
6.  
Click Start.  
Right-click Computer, and select Manage from the list.  
Expand System Tools.  
Expand Event Viewer.  
Expand Windows Logs.  
Select System.  
Windows Server environments 35  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
The following table lists common Event Log error messages, along with suggested solutions.  
Message  
Suggested solution  
To update the firmware, see "Upgrading the device firmware using Windows (on page  
23)."  
Error: ioDrive(x)  
firmware is too  
old. The firmware  
must be updated.  
1
2
3
Reinstall the Windows driver.  
Remove and reseat the IO Accelerator device.  
Remove the IO Accelerator and insert it in a different slot.  
Error: The  
ioDrive(x)  
initialization  
failed with error  
code 0xerrorcode  
(where errorcodeis a  
number which might  
vary)  
This error might appear after an unmanaged shutdown. You can use either the  
fio-attachcommand line utility or HP IO Manager to re-attach the IO Accelerator.  
This attach process might take up to ten minutes as the utility performs a consistency  
check on the drives.  
Error: The  
ioDrive(x)wasnot  
attached. Use the  
fio-attach  
utilitytorebuild  
the drive.  
The IO Accelerator must attach to the Windows operating system to be available to  
users and applications. This attach normally occurs at boot time. As part of this attach  
process, the IO Accelerator driver verifies the presence of an AutoAttachparameter  
in the Windows® registry. If you create this Registry parameter to disable auto-attach,  
the attach operation does not complete.  
Warning: The  
ioDrive(x)wasnot  
attached because  
auto-attach is  
disabled.  
To attach an unattached IO Accelerator:  
1
2
3
4
Run the HP IO Accelerator Management Tool.  
Select your unattached IO Accelerator from the Device Tree.  
Click Attach.  
Confirm the Attach operation.  
Your IO Accelerator now attaches to the Windows® operating system. To re-enable  
auto attach at boot time, see "Enabling auto attach ("Enabling Autoattach" on page  
38)."  
Windows Server environments 36  
Download from Www.Somanuals.com. All Manuals Search And Download.  
Maintenance  
Maintenance tools  
The IO Accelerator includes both software utilities to maintain the device and external LED indicators to  
display the status. You can also install SNMP as a monitoring option.  
The following are the most common tasks for maintaining your IO Accelerator. You can also use the IO  
Accelerator Management Tool application to perform firmware upgrades. For more information, see the HP  
IO Accelerator Management Tool User Guide.  
Common maintenance tasks  
The following tasks are the most common tasks for maintaining your IO Accelerator.  
NOTE: All commands require administrator privileges. Log in as root or use sudoto run the  
commands.  
Uninstalling the IO Accelerator driver package  
1.  
2.  
3.  
4.  
5.  
6.  
7.  
Go to Start > Control Panel.  
Click Administrative Tools.  
Click Computer Management.  
In the console tree at the left, click Device Manager.  
Expand HP IO Accelerators.  
Right-click an IO Accelerator.  
Click Uninstall. The Windows® operating system uninstalls the driver.  
Disabling AutoAttach  
When the IO Accelerator driver is installed, it is configured to automatically attach any devices when the  
driver is loaded. When necessary, disable the AutoAttach feature.  
To disable auto attach:  
1.  
Open the IO Accelerator Utilities folder. The default location is C:\Program Files\HP IO  
Accelerator\Utils\.  
2.  
3.  
Double-click the autoattachdisable.regfile.  
If necessary, confirm that you want to modify the registry. This creates a new DWORDparameter registry  
key called AutoAttach in  
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\fiodrive\Parameters.  
Your IO Accelerator does not automatically attach the next time you reboot the computer.  
Maintenance 37  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
4.  
When you finish troubleshooting the driver issue, use the IO Accelerator Management Tool to attach the  
IO Accelerators and make them available to Windows® operating system.  
Enabling Autoattach  
1.  
Open the IO Accelerator Utilities folder. The default location is C:\Program Files\HP IO  
Accelerator\Utils\.  
2.  
3.  
Double-click the autoattachenable.regfile.  
If necessary, confirm that you want to modify the registry. This action resets the AutoAttach  
parameter in the Registry. The next time that you reboot your Windows® operating system, your IO  
Accelerator automatically attaches.  
Unmanaged shutdown issues  
Unmanaged shutdowns, due to power loss or other circumstances, force the IO Accelerator to perform a  
consistency check during the reboot. This reboot might take several minutes or more to complete and is  
indicated by a progress percentage during a startup.  
You can cancel this consistency check by pressing Esc during the first 15 seconds after the Fusion-io  
Consistency Check message appears at the prompt. However, if you choose to cancel the check, the IO  
Accelerators remains unavailable to users until the check is completed. You can perform this check later using  
the IO Accelerator Management Tool Attach function.  
Although data written to the IO Accelerator is not lost due to unmanaged shutdowns, important data  
structures might not have been properly committed to the drive. This consistency check repairs these data  
structures.  
Enabling PCIe power override  
For PCIe IO Accelerators, if you have installed any dual IO Accelerator devices, such as the HP ioDrive2  
Duo, then the device might require additional power than the minimum 25 W provided by PCIe Gen2 slots  
to properly function. Even if additional power is not required for your device, all dual IO Accelerator devices  
that receive additional power might benefit with improved performance.  
HP ioDrive2 Duo devices must have additional power to properly function. For more information on which  
devices require additional power, see the HP PCIe IO Accelerator for ProLiant Servers Installation Guide.  
Additional power can be provided in two ways:  
External power cable—This cable ships with all dual ioMemory devices. For information on installing  
this cable, see the HP PCIe IO Accelerator for ProLiant Servers Installation Guide.  
NOTE: When a power cable is used, all of the power is drawn from the cable and no power is  
drawn from the PCIe slot.  
Enabling full slot power draw—Some PCIe slots provide additional power (often up to 75 W of power).  
If your PCIe slot is rated to provide at least 55 W, then you can allow the device to draw full power from  
the PCIe slot by setting a VSL module parameter. For more information on enabling this override  
parameter, see "Enabling the override parameter (on page 39)."  
Maintenance 38  
Download from Www.Somanuals.com. All Manuals Search And Download.  
     
CAUTION: If the PCIe slot is not capable of providing the needed amount of power, then  
enabling full power draw from the PCIe slot might result in malfunction or even damage to server  
hardware. The user is responsible for any damage to equipment due to improper use of the  
override parameter. HP expressly disclaims any liability for damage arising from improper use of  
the override parameter. To confirm the power limits and capabilities of each slot, as well as the  
entire system, contact the server manufacturer. For information about using the override  
parameter, contact HP Customer Support.  
NOTE: The override parameter overrides the setting that prevents devices from drawing more  
than 25 W from the PCIe slot. The parameter is enabled by device using the device serial  
numbers. Once the setting is overridden, each device might draw up to the full 55 W needed for  
peak performance.  
Before you enable the override parameter, ensure that each PCIe slot is rated to provide enough power  
for all slots, devices, and server accessories. To determine the power slot limits, consult the server  
documentation, BIOS interface, setup utility, or use the fio-pci-checkcommand.  
Important considerations  
If you are installing more than one dual IO Accelerator device and enabling the override parameter for  
each device, be sure the motherboard is rated to provide 55W power to each slot used. For example,  
some motherboards safely provide up to 75W to any one slot, but run into power constraints when  
multiple slots are used to provide that much power. Installing multiple devices in this situation might  
result in server hardware damage. Consult with the manufacturer to determine the total PCIe slot power  
available.  
The override parameter persists in the system and enables full power draw on an enabled device even  
if the device is removed and then placed in a different slot within the same system. If the device is placed  
in a slot that is not rated to provide 55W of power, your server hardware could experience a power  
drag.  
The override parameter is a setting for the IO Accelerator VSL software by server and is not stored in the  
device. When moved to a new server, the device defaults to the 25 W power limit until an external  
power cable is added or the override parameter is enabled for that device in the new server. To  
determine the total PCIe slot power available for the new server, consult the manufacturer.  
Enabling the override parameter  
1.  
Use one of the following methods to determine the serial number of each device to be installed in a  
compatible slot.  
o
Enter the fio-status command:  
Sample output:  
fio-status  
...  
Adapter: Dual Controller Adapter  
Fusion-io ioDrive2 DUO 2.41TB, Product Number:F01-001-2T41-CS-0001, FIO  
SN:1149D0969  
External Power: NOT connected  
PCIe Power limit threshold: 24.75W  
Connected ioMemory modules:  
Maintenance 39  
Download from Www.Somanuals.com. All Manuals Search And Download.  
   
fct2: SN:1149D0969-1121  
fct3: SN:1149D0969-1111  
In this example, 1149D0969 is the adapter serial number.  
If you have multiple IO Accelerator devices installed on your system, use the fio-beacon command  
to verify where each device is physically located.  
o
Inspect the adapter serial number labels on the IO Accelerator devices to determine the serial  
numbers. However, HP recommends confirming that each serial number is an adapter serial number  
by running the fio-status command. The adapter serial number label resides on the back of all HP  
ioDrive Duo devices and HP ioDrive2 Duo devices. On ioDrive Duos, the serial number is located on  
the PCB component that is attached to the PCIe connector.  
2.  
Set the module parameter for the fio-configcommand-line utility. For example:  
fio-config -p FIO_EXTERNAL_POWER_OVERRIDE <value>  
Where the <value> for this parameter is a comma-separated list of adapter serial numbers. For  
example:  
1149D0969,1159E0972,24589  
The -p option makes the parameter persistent after reboot.  
fio-config /dev/fct<x> -p PCIe_GLOBAL_SLOT_POWER_LIMIT 74750  
Where <x>is the device number for the device relative to the <value>specified adapter serial  
number. For example:  
fio-config /dev/fct2 –p PCIe_GLOBAL_SLOT_POWER_LIMIT 74760  
3.  
To enforce any parameter changes, you must reboot or unload and then load the drivers.  
Maintenance 40  
Download from Www.Somanuals.com. All Manuals Search And Download.  
Utilities  
Utilities reference  
The IO Accelerator installation packages include various command line utilities, installed by default to the  
C:\Program Files\HP IO Accelerator\Utilsfolder. These utilities provide a number of useful  
manners to access, test, and manipulate your device.  
Utility  
Purpose  
Makes an IO Accelerator available to the OS  
Lights the IO Accelerator external LEDs  
fio-attach  
fio-beacon  
fio-bugreport  
fio-config  
Prepares a detailed report for use in troubleshooting issues  
Enables configuration parameters for device operation  
Temporarily removes an IO Accelerator from OS access  
Performs a low-level format of an IO Accelerator  
fio-detach  
fio-format  
Searches for errors on the PCIe bus tree, specifically for  
PCIe IO Accelerators  
fio-pci-check  
Displays information about the device  
fio-status  
Securely removes information from tan IO Accelerator  
device  
fio-sure-erase  
Enables or disables the Trim feature  
Updates the IO Accelerator firmware  
fio-trim-config  
fio-update-iodrive  
NOTE: All utilities have –h (Help) and –v (Version) options.  
NOTE: To run these utilities from a command line using Windows Server®, you must either  
change directories to the directory that contains them (by default, C:\Program Files\HP IO  
Accelerator\Utils) or add that directory to your system path. See the documentation for  
your Windows® version for information about adding a directory to the system path.  
NOTE: The command-line utilities require administrator rights to run under Windows® operating  
systems. Right-click the Command Prompt menu item and select Run as administrator.  
fio-attach  
Description  
Attaches the IO Accelerator device and makes it available to Windows® operating systems. You can then  
partition the IO Accelerator device, or set it up as part of a RAID array, using the Windows® Disk  
Management utility. This command displays a progress bar and percentage as it completes the attach  
process.  
In most cases, the driver automatically attaches the device when loading and does a scan. You only need to  
run fio-attachif you ran fio-detachor if you set the auto_attach parameter to 0.  
Utilities 41  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
If the IO Accelerator device is in minimal mode, then auto-attachis disabled until the cause of the device  
being in minimal mode is fixed.  
Syntax  
fio-attach <device> [options]  
where <device>is the name given to the IO Accelerator during the installation process. The syntax is  
/dev/fctx, where xindicates the PCIe bus number where you installed the IO Accelerator. For example,  
the name /dev/fct4refers to the IO Accelerator device installed in PCIe Bus 4 in your Windows®  
operating system. Use the IO Accelerator Management Tool or fio-statusto view this bus number.  
You can specify multiple IO Accelerator devices. For example, /dev/fct /dev/fct2indicates the IO  
Accelerator devices installed in PCIe Buses 1 and 2 in your Windows® operating system.  
Options  
Option  
Description  
Attach only if clean.  
-c  
Quiet: disables the display of the progress bar and percentage.  
-q  
fio-beacon  
Description  
The fio-beacon utility turns on all three LEDs of the specified IO Accelerator for identification. You must  
detach the IO Accelerator device and then run fio-beacon.  
This utility is not applicable for the mezzanine format of the IO Accelerator, which has no externally visible  
LEDs.  
Syntax  
fio-beacon <device> [options]  
where <device>is the name given to the IO Accelerator during the installation process. The syntax is  
/dev/fctx, where xindicates the device number.  
Options  
Option  
Description  
Off: Turns off the three LEDs  
-0  
On (default): Lights the three LEDs  
-1  
-p  
Prints the PCI bus ID of the device at <device>to standard output. Usage and error information  
might be written to standard output rather than to standard error.  
fio-bugreport  
Description  
Prepares a detailed report of the device for use in troubleshooting issues.  
Utilities 42  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
This utility captures the current state of the device. When a performance or stability problem occurs with the  
device, run the fio-bugreportutility and contact HP support (http://www.hp.com/support) for  
assistance in troubleshooting.  
fio-bugreportruns several information-gathering utilities and combines the resulting data into a text file.  
The results are saved in the utils directory (default installation path is C:\Program Files\HP IO  
Accelerator\Utils\) in a .cabfile that indicates the date and time the utility was run.  
NOTE: If the utility recommends that you contact Fusion-io support, disregard that message and  
contact HP support (http://www.hp.com/support) instead.  
Syntax  
fio-bugreport  
Sample output  
C:\Users\<username>"\Program Files\HP IO Accelerator\Utils\fio-bugreport.exe"  
Generating bug report. Please wait, this may take a while...  
---------------------------------------------  
Gathering all Windows Event Logs...DONE  
Gathering Fusion-io Windows Event Logs...DONE  
Gathering System Information...DONE  
Running fio utilities...DONE  
Compressing to CAB file...DONE  
Bug report has successfully been created:  
fio-bugreport-20100222_192621.cab.  
Please e-mail this file to HP support  
For example, a bug report file named fiobugreport-20090921.192621.cabindicates the following:  
Date (20090921)  
Time (192621, or 19:26:21)  
fio-config  
Description  
Sets and gets configuration parameters for device operation. For a list of parameters, see "fio-config options  
(on page 44)."  
Syntax  
fio-config <device> [options]  
where <device>is the name given to the IO Accelerator during the installation process. The syntax is  
/dev/fctx, where xindicates the PCIe bus number where you installed the IO Accelerator. For example,  
the name /dev/fct4refers to the IO Accelerator device installed in PCIe Bus 4 in your Windows®  
operating system. Use the IO Accelerator Management Tool or fio-statusto view this bus number.  
Options  
Utilities 43  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Option  
Description  
Enumerates configuration parameter names and values.  
-e  
Gets the configuration parameter.  
Set and persist the configuration parameter.  
Set the configuration parameter in memory only.  
Print verbose information.  
-g <name>  
-p <name>  
-s <name>  
-V  
Print version information.  
-v  
fio-config options  
The following table describes the driver parameters you can set with the fio-configutility.  
IMPORTANT: fio-configoptions must be entered in uppercase to function properly.  
IMPORTANT: MSI (Message Signaled Interrupts) is enabled by default and it cannot be disabled  
using fio-config.  
Option  
Default  
(min/max)  
Description  
1 (0,1)  
Always attach the device on driver load (1).  
AUTO_ATTACH  
0 (0, 255)  
Interval (microseconds) to wait between hardware  
interrupts  
IODRIVE_TINTR_HW_WAIT  
No devices  
selected  
Allows selected devices to draw full power from the  
PCIe slot. Where the <value> for this parameter is a  
comma-separated list of adapter serial numbers.  
FIO_EXTERNAL_POWER_OVERRIDE  
0 (0,1)  
0 (0,1)  
0
Force minimal mode on the device (1), this  
parameter is set to false (0) by default.  
FORCE_MINIMAL_MODE  
PARALLEL_ATTACH  
Enable parallel attach of multiple devices (1), this  
parameter is set to false (0) by default.  
For the selected device, preallocate all memory  
necessary to have the drive usable as swap space.  
FIO_PREALLOCATE_MEMORY  
For example:  
fio-config /dev/fct0 -p  
FIO_PREALLOCATE_MEMORY =  
"1234,54321"  
where "1234" and "54321" are serial numbers  
obtained from fio-status.  
0 (affinity is  
enabled)  
WIN_DISABLE_ALL_AFFINITY  
When WIN_DISABLE_ALL_AFFINITYis set to 0,  
the driver will enable interrupt and worker thread  
affinity in the driver. When  
WIN_DISABLE_ALL_AFFINITYis set to 1, the  
driver will disable all affinity settings. This is an  
override of any other affinity settings. The driver  
must be reloaded for this parameter to take effect.  
0 (default;  
NUMA  
affinity is  
enabled)  
When WIN_DISABLE_DEFAULT_NUMA_AFFINITY  
is set to 0, during initialization, the driver will query  
the Windows® operating system for the affinity  
settings assigned to the adapter by the OS. This  
setting is known as the "default NUMA affinity".  
WIN_DISABLE_DEFAULT_NUMA_AFFINIT  
Y
Utilities 44  
Download from Www.Somanuals.com. All Manuals Search And Download.  
   
Option  
Default  
(min/max)  
Description  
After the affinity is queried correctly, the driver sets  
the affinity of the adapter interrupt and associated  
worker threads to the default operating system  
setting. This action has the effect of setting the  
affinity of the interrupt and worker threads to all  
processors on a single NUMA node in the system.  
When WIN_DISABLE_DEFAULT_NUMA_AFFINITY  
is set to 1, the driver will ignore the affinity settings  
assigned to the adapter by the OS. The driver must  
be reloaded for this parameter to take effect.  
N/A  
FIO_AFFINITY  
FIO_AFFINITYis a list of <affinity  
specification>triplets to specify the affinity  
settings of all adapters in the system. Each item in  
the triplet is separated by a comma, and each  
triplet set is separated by a semicolon. For more  
information on using this parameter, see "NUMA  
configuration (on page 67)."  
fio-detach  
Description  
Detaches and removes the corresponding /dev/fctxIO Accelerator block device. The fio-detach  
command waits until the device completes all read/write activity before executing the detach process. The  
command displays a progress bar and percentage as it completes the process.  
NOTE: Before using this utility, be sure that the device you want to detach is not currently  
mounted and in use.  
Syntax  
fio-detach <device> [options]  
where <device>is the name given to the IO Accelerator during the installation process. The syntax is  
/dev/fctx, where xindicates the PCIe bus number where you installed the IO Accelerator. For example,  
the name /dev/fct4refers to the IO Accelerator device installed in PCIe Bus 4 in your Windows®  
operating system. To view this bus number, use the IO Accelerator Management Tool or fio-status.  
You can specify multiple IO Accelerator devices. For example, /dev/fct1 /dev/fct2indicates the  
devices installed in PCIe Buses 1 and 2 in your Windows® operating system.  
Options  
Option  
Description  
Force: Causes an immediate detach (does not save metadata).  
-f  
Immediate: Causes a forced immediate detach (does not save metadata). This will fail if the device  
is in use by the OS.  
-i  
-q  
Quiet: Disables the display of the progress bar and percentage.  
Utilities 45  
Download from Www.Somanuals.com. All Manuals Search And Download.  
   
IMPORTANT: Although the -f(force) option causes the IO Accelerator device to detach, even in  
a RAID setup, it is strongly recommended to take the drives/volume offline using the Windows®  
Disk Management plug-in and then perform the detach. Forcing the detach may result in loss of  
data.  
Notes  
Attempting to detach an IO Accelerator device might fail with an error indicating that the device is busy. This  
might occur if the device is part of a software RAID (0,1,5) volume. The Windows® operating system refuses  
the request to detach the drive associated with the IO Accelerator device because it is part of a RAID volume  
and might cause the volume to fail. This does not occur with simple volumes (such as a single IO Accelerator  
device). To detach a simple volume, take the volume offline using the Disk Management MMC plug-in  
application.  
fio-format  
Description  
The fio-formatcommand performs a low-level format of the IO Accelerator. This format is different from  
a format performed by the operating system. The utility displays a progress bar and percentage as it  
completes the format.  
The IO Accelerator ships pre-formatted, so you only have to run the fio-formatcommand if you must  
change the logical size of the device.  
CAUTION: Use this utility with care since it deletes all user information on the IO Accelerator.  
Syntax  
fio-format <device> [options]<device>  
where <device>is the name given to the IO Accelerator during the installation process. The syntax is  
/dev/fctx, where xindicates the PCIe bus number where you installed the IO Accelerator. For example,  
the name /dev/fct4refers to the IO Accelerator device installed in PCIe Bus 4 in your Windows®  
operating system. To view this bus number, use the IO Accelerator Management Tool or fio-status.  
Options  
Option  
Description  
Set the block (sector) size, in bytes or KiBytes (base 2). The default is 512 bytes. For  
example: -b 512B or -b 4K (B in 512B is optional).  
-b <size  
B|K>  
Force the format size, bypassing normal checks and warnings. This option might be  
needed when fio-formatdoes not proceed properly. The "Are you sure?" prompt still  
appears unless you use the -yoption. This option can only be used with the -ooption.  
-f  
Quiet mode: Disable the display of the progress-percentage indicator.  
-q  
Set the device capacity as a specific size (in TB, GB, or MB) or as a percentage of the  
advertised capacity:  
-s <size  
M|G|T|%>  
TNumber of terabytes (TB) to format  
GNumber of gigabytes (GB) to format  
MNumber of megabytes (MB) to format  
%Percentage, such as 70% (the percent sign must be included)  
Disable fast rescan on unclean shutdowns at the cost of some capacity.  
-R  
Utilities 46  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Option  
Description  
Auto-answer "yes" to all queries from the application (bypass prompts).  
-y  
fio-pci-check  
Description  
Checks for errors on the PCI bus tree, specifically for IO Accelerators. This utility displays the current status of  
each IO Accelerator. It also prints the standard PCI Express error information and resets the state.  
It is normal to see a few errors (perhaps as many as five) when fio-pci-checkis initially run. Subsequent  
runs should reveal only one or two errors during several hours of operation.  
The IO Accelerator driver must be loaded to run this utility. Some PCI errors cannot be reset in your  
Windows® operating system.  
Syntax  
fio-pci-check [options]  
Options  
Option  
Description  
1 = Disable the link; 0 = bring the link up  
-d <value>  
Scan every device in the system.  
-f  
-r  
-v  
-y  
Force the link to retrain.  
Verbose: Print extra data about the hardware.  
"Yes" is forced when the user is asked to continue.  
fio-status  
Description  
The fio-statuscommand provides detailed information about the specified or all installed IO  
Accelerators, either fctxor fioxdevices. The fio-statuscommand must run with root privileges and  
requires that the driver be loaded. If it finds no device, it displays an index number instead.  
fio-statusprovides alerts for certain error modes, such as a minimal-mode, read-only mode, and  
write-reduced mode, describing what is causing the condition.  
Syntax  
fio-status [<device>] [options]  
where <device>is the name given to the IO Accelerator during the installation process. The syntax is  
/dev/fctx, where xindicates the PCIe bus number where you installed the IO Accelerator. For example,  
the name /dev/fct4refers to the IO Accelerator device installed in PCIe Bus 4 in your Windows®  
operating system. To view this bus number, use the IO Accelerator Management Tool or fio-status.  
If you do not specify a name, fio-statusdisplays information on all installed IO Accelerators. If the IO  
Accelerator is not loaded, this parameter is ignored.  
Options  
Utilities 47  
Download from Www.Somanuals.com. All Manuals Search And Download.  
   
Option  
Description  
Report all available information for each device.  
-a  
Show all errors and warnings for each device. This option is for diagnosing issues, and it hides  
other information such as format sizes.  
-e  
Count: Report only the number of ioMemory devices installed.  
-c  
-d  
Show basic information set plus the total amount of data read and written (lifetime data  
volumes). This option is not necessary when the -aoption is used.  
Format JSON: creates the output in JSON format.  
-fj  
Format XML: creates the output in XML format.  
-fx  
-u  
Show unavailable fields. Only valid with -fjor -fx.  
Show unavailable fields and details why. Only valid with -fjor -fx.  
-U  
Print the value for a single field (see the next option for field names). Requires that a device be  
-F<field>  
specified. Multiple -Foptions may be specified.  
-l  
List the fields that can be individually accessed with -F.  
Starting with version 3.0.0 and later, the standard formatting of the fio-statusoutput changed. This  
change affects any custom management tools that used the output of this utility.  
Basic Information: If no options are used, fio-statusreports the following basic information:  
Number and type of cards installed in the system  
Software version  
Adapter information:  
Adapter type  
Product number  
External Power Supply  
PCI power limit threshold (if available)  
Connected IO Accelerator devices  
Block device information:  
Attach status  
Product name  
Product number  
Serial number  
PCIe slot number  
Firmware version  
Size of the device, out of total capacity  
Internal temperature (average and maximum, since IO Accelerator load) in degrees Centigrade  
Health status: healthy, nearing wearout, write-reduced or read-only  
Reserve capacity (percentage)  
Warning capacity threshold (percentage)  
Utilities 48  
Download from Www.Somanuals.com. All Manuals Search And Download.  
Data Volume Information: If the -doption is used, then the following data volume information is reported in  
addition to the basic information:  
Physical bytes written  
Physical bytes read  
All Information: If the -aoption is used, then all information is printed, which includes the following  
information in addition to basic and data volume information:  
Adapter information:  
Manufacturer number  
Date of manufacture  
Power loss protection status  
PCIe bus voltage (avg, min, max)  
PCIe bus current (avg, max)  
PCIe power limit threshold (watts)  
PCIe slot available power (watts)  
PCIe negotiated link information (lanes and throughput)  
Block device information:  
Part number  
Manufacturer's code  
Manufacturing date  
Vendor and sub-vendor information  
Size of the device, out of total capacity  
Format status and sector information (if device is attached)  
FPGA ID and Low-level format GUID  
PCIe slot available power  
PCIe negotiated link information  
Card temperature, in degrees Centigrade  
Internal voltage: avg. and max.  
Auxiliary voltage: avg. and max.  
Percentage of good blocks, data and metadata  
Lifetime data volume statistics  
Error Mode Information: If the IO Accelerator is in minimal mode, read-only mode, or write-reduced mode  
when fio-statusis run, the following differences occur in the output:  
Attach status is Status unknown: Driver is in MINIMAL MODE:  
The reason for the minimal mode state is displayed (such as Firmware is out of date. Update  
firmware.)  
Geometry and capacity information not available.appears.  
No media health information appears.  
Utilities 49  
Download from Www.Somanuals.com. All Manuals Search And Download.  
fio-sure-erase  
CAUTION: Before using this utility, back up any important data.  
IMPORTANT: As a best practice, do not use this utility if you have any IO Accelerator devices  
installed in the system that you do not want to clear or purge. First remove any devices that you  
do not want to accidentally erase. After the data is removed with this utility it is not recoverable.  
IMPORTANT: If the device is in Read-only mode, perform a format using fio-formatbefore  
running fio-sure-erase. If the device is in Minimal mode, then fio-sure-erasecannot  
erase the device. Updating the firmware may take the device out of Minimal Mode. If the device  
remains in Minimal mode, contact HP Customer Support for further assistance.  
IMPORTANT: After using fio-sure-erase, format the device using fio-formatbefore  
using the device again.  
To run the fio-sure-eraseutility, the block device must be detached. See fio-detachfor more  
information.  
Description  
The fio-sure-eraseis a command-line utility that securely removes data from IO Accelerator devices. It  
complies with the Clear and Purge level of destruction from the following standards:  
DOD 5220.22-M–Comply with instructions for Flash EPROM  
NIST SP800-88–Comply with instructions for Flash EPROM  
Registry requirement  
To create a registry key to configure the driver for ECC-bypass mode:  
1.  
Locate the following key:  
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\fiodrive\Parameters  
2.  
3.  
Create a DWORD key underneath it called BypassECCand set the value to 1.  
Restart the computer before running the utility.  
Syntax  
fio-sure-erase [options] <device>  
where <device>is the name given to the IO Accelerator during the installation process. The syntax is  
/dev/fctx, where xindicates the PCIe bus number where you installed the IO Accelerator. For example,  
the name /dev/fct4refers to the IO Accelerator device installed in PCIe Bus 4 in your Windows®  
operating system. To view this bus number, use the IO Accelerator Management Tool or fio-status.  
The fio-sure-erasesyntax works on individual IO Accelerator devices. For example, if you are planning  
to purge an IO Accelerator Duo device, you must perform this operation on each of the IO Accelerator  
devices.  
Options  
If fio-sure-eraseis run with no options, a Clear operation is performed.  
Utilities 50  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Option  
Description  
Purge instead of Clear: Performs a write followed by an erase. Purging the device might take  
several hours to complete, depending on the size of the device.  
-p  
No confirmation: Does not require a yes/no response to execute the utility.  
Quiet: Do not display the status bar.  
-y  
-q  
Clear support  
A Clear is the default state of running fio-sure-erase(with no options), and is the act of performing full  
low-level erase (every cell pushed to 1) of the entire NAND media, including retired erase blocks.  
Metadata that is required for operation will not be destroyed (media event log, erase counts, physical bytes  
read/written, performance and thermal history), but any user-specific metadata will be destroyed.  
A Clear operation performs the following:  
1.  
Creates a unity map of every addressable block (this allows fio-sure-eraseto address every block,  
including previously unmapped bad blocks).  
2.  
3.  
4.  
Performs an erase cycle for each block (every cell is pushed to 1).  
Restores the bad block map.  
Formats the device to make the device usable again. The utility erases all of the headers during the  
clear.  
Purge support  
A Purge operation is implemented by using the -poption with fio-sure-erase. Purge is the act of first  
overwriting the entire NAND media (including retired erase blocks) with a single character (every cell written  
to logical 0), and then performing a full chip erase (every cell pushed to 1) across all media (including retired  
erase blocks).  
Metadata that is required for operation will not be destroyed (media event log, erase counts, physical bytes  
read/written, performance and thermal history), but any user-specific metadata will be destroyed.  
A Purge operation performs the following:  
1.  
Creates a unity map of every addressable block (this allows fio-sure-eraseto address every block,  
including previously unmapped bad blocks).  
2.  
3.  
4.  
5.  
Performs a write cycle for each block (every cell written to 0).  
Performs an erase cycle for each block (every cell pushed to 1).  
Restores the bad block map.  
Formats the drive to make the drive usable again. The utility erases all of the headers during the clear.  
fio-trim-config  
Description  
Enables or disables the Trim feature, which reclaims available space from the file system. It is generally  
recommended that Trim be enabled for best performance. TRIM is enabled by default. Running this utility with  
no options displays the current Trim status. For more information, see "Trim support (on page 57)."  
NOTE: This utility affects all IO Accelerators in the system. You cannot enable or disable Trim for  
only selected devices.  
Utilities 51  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Syntax  
fio-trim-config [options]  
Options  
Option  
-d  
Description  
Disable Trim on this computer.  
(Default) Enable Trim on this computer.  
-e  
fio-update-iodrive  
Description  
The fio-update-iodriveutility updates the HP IO Accelerator firmware. This utility scans the PCI-Express  
bus for all HP IO Accelerators and updates them all sequentially.  
You must detach all IO Accelerator devices before updating the firmware. Detach the device from the system  
using the fio-detach utility. See "fio-detach (on page 45)" for more information.  
To update one or more specific drives (rather than all drives):  
Make sure the IO Accelerator driver is loaded.  
Use the -dor -soption.  
CAUTION: Do not turn off the power during a firmware upgrade, because this might cause  
device failure. If a UPS is not in place, consider adding one to the system before performing a  
firmware upgrade.  
CAUTION: Note that when running multiple upgrades in sequence (such as going from driver  
version 1.2.7 to 2.1.0 to 2.3.1), it is important to load the IO Accelerator driver after each  
firmware upgrade step. Otherwise the on-drive format will not be changed, and there will be data  
loss.  
CAUTION: Do not use this utility to downgrade the IO Accelerator to an earlier version of the  
firmware. Doing so might result in data loss. If you need to downgrade your firmware, contact HP  
customer support. For more information, see "Technical support."  
NOTE: If you receive an error message when updating the firmware instructing you to update the  
midprom information, contact Customer Support. For more information, see "Technical support."  
The default action (without using the -dor -soption) is to upgrade all IO Accelerators with the firmware  
contained in the <iodrive_version.fff>file. Confirm that all devices need the upgrade prior to  
running the update. If in doubt, use the -p(Pretend) option to view the possible results of the update.  
Syntax  
fio-update-iodrive <iodrive_version.fff> [options]  
where <iodrive_version.fff>is the path and firmware archive file provided by HP. This is typically  
C:\Program Files\HP IO Accelerator\Firmware. This parameter is required.  
Option  
Description  
-d  
Updates the specified devices (by fctx, where x is the number of the device shown in  
fio-status). If this option is not specified, all devices are updated.  
Utilities 52  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Option  
Description  
Force upgrade. Used primarily to downgrade to an earlier firmware version. If the IO Accelerator  
-f  
driver is not loaded, this option also requires the -doption.  
Lists firmware available in the archive.  
-l  
-c  
-p  
-q  
-y  
-s  
Clears locks placed on a device.  
Pretend. Shows what updates are done. The firmware is not modified.  
Runs the update process without displaying the progress bar or percentage.  
Confirm all warning messages.  
Updates devices in selected slots, using ‘*as a wildcard for boards. The slots are in the following  
format: [[[[<domain>]:]<bus>]:][<slot>][.[<func>]]  
All three external LED indicators light up during the update process.  
CAUTION: Use the -foption with care because it might damage your board.  
CAUTION: Use the -dand -soptions with care because updating a device in the wrong slot  
might damage it.  
Utilities 53  
Download from Www.Somanuals.com. All Manuals Search And Download.  
Monitoring IO Accelerator health  
NAND flash and component failure  
The IO Accelerator is a highly fault-tolerant storage subsystem that provides many levels of protection against  
component failure and the loss nature of solid state storage. However, as in all storage subsystems,  
component failures might occur.  
When a large enough number of data blocks is retired due to error, the NAND flash media is considered  
worn out. By properly selecting NAND flash media for the hosted application and proactively monitoring  
device age and health, you can assure reliable performance over the intended product life.  
Health metrics  
The IO Accelerator driver manages block retirement using pre-determined retirement thresholds. The IO  
Accelerator Management Tool and the fio-statusutility show a health indicator that starts at 100 and  
counts down to 0. As certain thresholds are crossed, various actions are taken.  
At the 10% healthy threshold, a one-time warning is issued. For methods for capturing this alarm event, see  
"Health monitoring techniques (on page 54)."  
At 0%, the device is considered unhealthy. It enters write-reduced mode, which somewhat prolongs its  
lifespan so data can be safely migrated. In this state, the IO Accelerator behaves normally except for the  
reduced write performance.  
At some point after the 0% threshold, the device enters read-only mode. Any attempt to write to the IO  
Accelerator causes an error. Some file systems might require special mount options to mount a read-only  
block device, beyond specifying that the mount should be read-only. For example, under Linux, ext3  
requires that -oro,noloadis used. The noloadoption tells the file system not to try to replay the journal.  
Read-only mode should be considered a final opportunity to migrate data off the device since device failure  
is more likely with continued use.  
The IO Accelerator might enter failure mode. In this case, the device is offline and inaccessible. This can be  
caused by an internal catastrophic failure, improper firmware upgrade procedures, or device wears out.  
Health monitoring techniques  
fio-status  
Output from the fio-statusutility shows the health percentage and drive state. These items are in bold in  
the following sampleoutput.  
Found 1 ioDrive in this system  
Fusion-io driver version: 2.1.0 build 19032  
Adapter: ioDrive  
HP 160GB SLC PCIe ioDrive for ProLiant Servers, Product Number:600278-B21  
Monitoring IO Accelerator health 54  
Download from Www.Somanuals.com. All Manuals Search And Download.  
   
...  
Media status: Healthy; Reserves: 100.00%, warn at 10.00%; Data: 99.12%  
Lifetime data volumes:  
...Physical bytes written: 6,423,563,326,064  
...Physical bytes read: 5,509,006,756,312  
The following Health Status messages are produced by the fio-statusutility:  
Healthy  
Low metadata  
Read-only  
Reduced-write  
Unknown  
HP IO Accelerator Management Tool  
In the Device Report tab, look for the Reserve Space percentage in the right column. The higher the  
percentage, the healthier the drive is likely to be.  
SNMP  
For details on how to configure SNMP health indicators, see "Setting up SNMP for Windows operating  
systems (on page 29)."  
Flashback substitution events  
The IO Accelerator is equipped with Flashback substitution capability. This capability allows for predictive  
NAND flash failures to occur and keep the IO Accelerator functioning by substituting out bad locations or  
even a whole NAND chip. The output of the fio-status –acommand will show if any flashback  
substitution events have occurred.  
The following is an example of the output after entering the fio-status –acommand. The numbers  
represent the bank (0-3) and chip number that was substituted.  
First flashback event -> No action to take as infant mortality condition on NAND  
chip was caught per design.  
Second flashback event –> Run fio-bugreport. Return the card (if under warranty)  
when you see two or more flashback substitution events.  
Flashback active on (bank a/chip m) and (bank b/chip n) (multi-bank failure)  
Flashback active on (bank a/chip m) and (bank b/chip m) (chip failure)  
On the second flashback event, include the output from running the fio-bugreportcommand when you  
return the card.  
Software RAID and health monitoring  
Software RAID stacks are typically designed to detect and mitigate the failure modes of traditional storage  
media. The IO Accelerator attempts to fail as gracefully as possible, and its new failure mechanisms are  
compatible with existing software RAID stacks. A drive in write-reduced mode participating in a write-heavy  
workload is evicted from a RAID group for failure to receive data at a sufficient rate. A drive in read-only  
Monitoring IO Accelerator health 55  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
mode is evicted when write I/Os are returned from the device as failed. Catastrophic failures are detected  
and handled just as though they were on traditional storage devices.  
Monitoring IO Accelerator health 56  
Download from Www.Somanuals.com. All Manuals Search And Download.  
Trim support  
Introduction to Trim  
Trim is used to address a unique property of solid state drives. The issue occurs when a user deletes a file, and  
then the drive that contains the file does not recognize that it can reclaim that space. Instead, the drive reads  
the data as valid until the system informs the drive it can overwrite that data. With a typical hard drive you  
can continually write to the same sector of the drive without significant degradation. However, a solid state  
drive writes to different flash memory areas when the same logical sector is written to.  
Because of the way flash memory handles writes, having a full SSD can cause significant reduction in write  
throughput. Currently available storage stacks and file systems were not designed with this caveat in mind.  
The SSD software must assume that all sectors contain valid data, and therefore treat the drive as full. In  
reality, a file system does not normally contain a full drive of data.  
Trim is an enhancement to existing file systems that informs the SSD software of logical sectors that do not  
contain valid user data. Trim retrieves this information from a file system when a file has been deleted, and  
it informs the drive that it can reclaim the space held by that file. This feature enables the wear-leveling  
software to reclaim that space as reserve to handle future write operations.  
For Trim to perform this task, it continually runs in the background and monitors the file systems being used  
on IO Accelerator devices. The process is meant to run in the background so as not to noticeably impact the  
performance of the computer.  
Trim platforms  
The Trim feature is available on IO Accelerators for Windows® operating systems starting with Windows®  
XP and Windows Server® 2003.  
NOTE: Starting with Windows® 7 and Windows Server® 2008 R2, Trim capabilities are built  
into the operating system. IO Accelerators support Windows® 7 Trim, which means that IO  
Accelerator Trim is unnecessary under Windows® 7 and Windows Server® 2008 R2, and it will  
not run when either of these is present. No system changes are needed. The IO Accelerator Trim  
service automatically shuts off if it is attempting to run on a Windows® 7 or Windows Server®  
2008 R2 or later operating system.  
Using Trim  
Trim runs quietly in the background as a Windows® Service a few minutes after the computer is started, so  
as not to slow down the boot process. However, you can also fine-tune Trim by using the configurable  
settings.  
Trim support 57  
Download from Www.Somanuals.com. All Manuals Search And Download.  
   
Starting and stopping Trim  
To start or stop the Trim service through the Windows® Services Manager:  
1.  
2.  
Search for IO Accelerator Trim Service.  
Click Start service or Stop service.  
Stopping the service only stops it for that session. Trim begins running again when the computer is restarted.  
Enabling Trim  
1.  
2.  
Open a registry editor.  
Navigate to  
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\fiodrive\Parameters  
3.  
Add the TrimEnabledvalue as a DWORD (see TrimEnabled below). Stop and restart the Trim service,  
or restart the computer.  
TrimEnabled  
This registry value controls whether Trim runs, regardless of the user starting and stopping Trim. If this value  
is present and set to 1, then Trim runs normally (unless you have stopped the service). If this value is set to 0,  
then Trim does not run, regardless of whether you have started the service. If this value is not present, Trim  
runs unless you stop it. The TrimEnabledvalue persists across sessions, so if you restart the computer and  
set this value to 0, Trim does not run when the computer is restarted.  
Controlling Trim aggressiveness  
You can control how aggressive the Trim service is. The default settings for Trim minimize impact on computer  
performance while enabling the IO Accelerators to perform optimally. However, if a drive is being used with  
a write-intensive application, then you can set Trim to be more aggressive. To do this, you must add registry  
entries.  
1.  
2.  
Open a registry editor.  
Navigate to  
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\fiodrive\Parameters  
3.  
4.  
5.  
Add the TrimTimeDelayInSecondsvalue as a DWORD (see the TrimTimeDelayInSeconds  
description below).  
Add the BitmapRetrievalDelayInMillisecondsvalue as a DWORD (see the  
BitmapRetrievalDelayInMilliseconds description below).  
Stop and restart the Trim service, or restart the computer.  
TrimTimeDelayInSeconds  
This registry value controls how quickly Trim starts running again after it has finished processing all the  
volumes on a system. For example, if this value is set to 600 (10 minutes in seconds), then Trim will process  
all the volumes on a system, sleep for 10 minutes, and then process the volumes again and continue this  
cycle. This value can be any number between 0 and 4,294,967,295. If this value is not set or present in the  
registry, then Trim assumes a default of 900 seconds (or 15 minutes).  
BitmapRetrievalDelayInMilliseconds  
This value affects how much impact Trim has on the CPU. The default is 25 milliseconds but can be any  
number between 0 (most aggressive) and 4,294,967,295 (not aggressive at all).  
Trim support 58  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Trim configurations  
Trim can be used with the following configurations and features:  
RAID volumes (mirrored, spanned, or striped).  
NOTE: RAID 5 (striped with parity) is not currently supported.  
Simple volumes (no RAID).  
Any combination of the above RAID levels across multiple devices, so long as there is at least one IO  
Accelerator in the RAID set.  
Multiple partitions on the same IO Accelerator device.  
NTFS and FAT32 file systems.  
Volumes with mount points (no <drive letter>is defined).  
Compressed volumes.  
Different cluster sizes, packet sizes, and sector sizes.  
Extended and shrunk volumes.  
Trim support 59  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Using Windows page files with the IO  
Accelerator  
Introduction to Windows page files  
This section describes how to effectively use paging (swap) files on IO Accelerator devices with Windows®  
operating systems.  
Using a page file with a traditional disk drive places practical limits on the usable size of the page file and  
virtual memory due to the poor performance of disk drives in relation to RAM. Placing the operating system  
paging file on one or more IO Accelerators enables much larger page files and usable virtual memory. This  
is due to the much faster response times and bandwidth on IO Accelerators versus hard disks. IO Accelerator  
software prior to version 2.0.2 does not support paging files.  
Configuring IO Accelerator paging support  
The IO Accelerator 2.0.2 driver and later versions can be configured to support paging files on one or more  
IO Accelerators. Configuring paging support requires that each IO Accelerator used with a paging file  
preallocates the worst-case amount of memory it might need in any possible I/O scenario. Memory is  
pre-allocated on a per-adapter (ioDIMM) instance.  
Because of the extra host RAM memory use, paging must be enabled only on IO Accelerators that actually  
hold a paging file. You can place a single paging file on more than one IO Accelerator. Windows®  
operating systems stripe paging I/O across all available paging files, providing additional performance to  
the VM subsystem.  
RAM consumption  
The amount of RAM preallocated per IO Accelerator depends on the total size of the device and the sector  
(block) size selected when formatting the drive (with fio-format). The following table lists the estimated  
memory consumption for the various device and sector sizes.  
IO Accelerator size (GB)  
512 byte sector  
2,278  
4096 byte sector  
284  
80  
4,556  
569  
160  
320  
640  
9,132  
1,141  
2,282  
18,264  
Host RAM Consumption in Mebibytes (MiB); 1024 x 1024 = 1 MiB  
Using a larger sector size significantly reduces the amount of host memory consumption needed for paging  
support. HP recommends using a 4K sector size be used because:  
that is generally the natural size of a host memory page, and  
it minimizes overall host memory consumption.  
Using Windows page files with the IO Accelerator 60  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
On Windows® operating systems, NTFS will generally use a cluster size of 4K, so formatting to 512 is not  
useful except for applications that are compatible only with 512-byte sector sizes (such as Windows® XP  
and Windows Server® 2003).  
The indicated amount is needed per IO Accelerator that supports paging. You must carefully plan which IO  
Accelerators are used to hold a paging file.  
Non-paged memory pool  
Pre-allocated memory for the IO Accelerator comes from the Windows kernel non-paged memory pool. This  
pool dynamically grows as system components consume additional kernel memory. The maximum size of this  
pool is restricted as follows:  
Windows Server 2003, 2008 R1/R2, 2012—75% of RAM up to a maximum of 128GB.  
Windows Vista and Windows 7—40% of RAM up to a maximum of 128GB.  
The amount of in-use non-paged pool memory should be noted when planning page file usage. This is  
because the IO Accelerator preallocates RAM and that reduces the available physical non-paged memory.  
The driver fails to load if the total preallocated memory plus the in-use non-paged memory exceeds the  
maximum non-paged memory pool.  
To determine the total non-paged memory pool use for two IO Accelerators, use the following example:  
One IO Accelerator is 80GB and the other is 160GB  
Both are formatted with a 4K sector size  
Both will support paging files  
The current allocated non-paged pool is obtained from Task Manager and, in this example, has a value of  
576 MiB. (Values shown in Task Manager are in MiB (1024x1024 = 1 MiB)). The total RAM on the system  
is 8,196 MiB and the operating is Windows Server 2008 R2.  
To calculate the total available non-paged pool, use the following formula:  
(8196 MiB x 0.75) - 576 - 284 (80GB IO Accelerator) - 569 (160GB IO Accelerator)  
This still leaves 4,718 MiB available for non-paged pool.  
Enabling and disabling paging support  
Memory preallocation occurs during IO Accelerator driver initialization. To enable paging support, you must  
enable the FIO_PREALLOCATE_MEMORY configuration item. This can be done using the fio-config  
command-line utility. This parameter is assigned a string with a list of decimal serial numbers of the IO  
Accelerators that will support a paging file. The driver performs memory preallocation for those instances.  
The following example illustrates using the fio-configutility, which enables paging and preallocation on  
two IO Accelerators with serial numbers 1234 and 17834. Serial number information can be obtained using  
the fio-statusutility.  
fio-config -p FIO_PREALLOCATE_MEMORY "1234,17834"  
To disable paging support on all drives, use a value of 0for FIO_PREALLOCATE_MEMORY:  
fio-config -p FIO_PREALLOCATE_MEMORY "0"  
To query the current value, run this command:  
fio-config -g FIO_PREALLOCATE_MEMORY  
Using Windows page files with the IO Accelerator 61  
Download from Www.Somanuals.com. All Manuals Search And Download.  
   
An alternate method to manage (enable or disable) paging support is to use the IO Accelerator Management  
Tool.  
NOTE: You must reload the IO Accelerator driver for the new preallocation setting to take effect.  
Typically this can be done by restarting the machine or using disable/enable within Device  
Manager for each IO Accelerator instance.  
Using the Windows System Properties to change paging file configuration requires a system  
restart before the properties are applied. Therefore, you can change both  
FIO_PREALLOCATE_MEMORY and the system page file configuration and then apply both with  
a single restart.  
Windows page file management  
By default, the IO Accelerator driver disables support for page files. The previous section described how to  
enable support for page files on one or more IO Accelerators. The following section describes how to work  
with the built-in Windows® control panels to configure and set up paging files on IO Accelerators.  
Setting up page files  
To set up page files in Windows® operating systems:  
1.  
2.  
3.  
4.  
5.  
In the Control Panel, double-click System.  
On the Task pane, click Advanced system settings.  
On the Advanced tab, click Settings. The Performance Options dialog box opens.  
On the Advanced tab, click Change. The Virtual Memory dialog box opens.  
Using this dialog box, configure a page file for each available drive in the system. Selecting the  
Automatically manage paging file size for all drives checkbox causes Windows® operating systems to  
create a single page file on the system drive, which is the drive the operating system starts from. This  
checkbox should be cleared when using an IO Accelerator with a paging file.  
Windows® operating systems support up to 16 distinct paging files. To enable a page file on an IO  
Accelerator:  
1.  
2.  
3.  
4.  
5.  
6.  
Choose the IO Accelerator from the device list.  
Select the Custom size button.  
Provide values in the Initial size and Maximum size fields.  
Click Set to save the setting. Do not omit this step, or your changes will be lost.  
Click OK.  
When prompted to restart, click Yes. This is necessary for the new page file settings to take effect.  
To remove a paging file on the drive, follow the steps above but select No paging file. For performance  
reasons, typically you will remove all paging files on any system hard disk.  
NOTE: The Virtual Memory dialog box enables pages files to be configured on available IO  
Accelerators, even if the IO Accelerator has not been configured to support a page file. Even  
though the dialog box enables the page file, following the required restart, no page file is created  
on the device. Follow the directions in "Enabling and disabling paging support (on page 61)" to  
properly enable page file support on one or more IO Accelerators.  
Using Windows page files with the IO Accelerator 62  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
System drive paging file configuration  
By default, Windows® operating systems create and manage a page file on the system boot drive (typically  
a hard disk), which is where the operating system is installed. Maintaining a regular page file on the system  
hard disk is typically not optimal, because the hard disk I/O performance is many orders of magnitude  
slower than an IO Accelerator. To resolve this issue, you can eliminate or minimize the size of the system boot  
drive page file, as explained later. Enabling page files on IO Accelerators but not the system drive improves  
VM subsystem performance since the VM manager stripes I/O across all available page files. Additionally,  
the IO Accelerators act as a large memory store, which can greatly improve memory usage for large  
applications.  
The Windows® kernel uses the system disk page file to store crash dumps. Crash dumps might be small  
(mini-dumps) or large (full-kernel memory dumps). Typically, running without dump file support or with a small  
dump file is adequate. There are several possible system drive page file configurations:  
Eliminate all page files on any hard disks, including the system boot drive. Although this maximizes  
paging I/O on IO Accelerators, no post-mortem crash dump file will be available if a system crash  
occurs. However, you might be able to re-enable a page file on the system drive and then reproduce the  
crash scenario.  
Create a minimal-size page file on the system boot drive. The recommended minimum size is 16MB,  
although the operating system might warn that a minimum 400MB page file is needed.  
Create a page file large enough for a full-kernel memory dump. This typically requires a page file at  
least the size of installed RAM, with some recommending the size equal to RAM x 1.5.  
To view or change the crash dump configuration:  
1.  
2.  
3.  
4.  
Go to the System Properties dialog box.  
Click the Advanced tab.  
In the Startup and Recovery section, click Settings. The Startup and Recovery dialog box opens.  
In the System Failure section, you can change settings to handle the system log, restart, and debugging  
information.  
Guaranteeing minimum committable memory  
If you enable System managed size or set a Custom size in the Virtual Memory dialog box, do so with care.  
If the initial size is less than the desired amount of committable virtual memory, this size might cause an  
application to have memory allocation failures if the amount of committed memory exceeds the currently  
allocated page file size or the initial size value. When committed memory exceeds the current page file size,  
a request to allocate additional memory will fail. The Windows Virtual Memory manager will slowly increase  
the size of the paging file up to the available size of its drive or to the Maximum size custom setting,  
whichever is smaller.  
If you want to use a large amount of committed virtual memory (more than 1.5 times the amount of RAM) and  
avoid application memory allocation errors, the initial and maximum committed memory must be explicitly  
set for the expected application committed memory usage. These values should generally be the same.  
The following articles explain in detail how to size the page file appropriately.  
Main Article Link: Pushing the Limits of Windows  
Specific section that documents virtual memory: Pushing the Limits of Windows: Virtual Memory  
Using Windows page files with the IO Accelerator 63  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Verifying page file operation  
To verify that a page file is actively placed on an IO Accelerator, you can browse for hidden files at the drive  
root. For example, run the following command at a prompt:  
dir c: /ah  
In the output listing there should be a file called pagefile.sys. If no page file is present, recheck the page  
file configuration in the Virtual Memory dialog box to verify that page file support has been enabled on the  
queried IO Accelerator.  
Virtual Memory performance  
Using the IO Accelerator as the paging store can improve overall Virtual Memory system performance.  
Actual benefits will vary widely with the application virtual memory usage and with hardware platform and  
performance.  
Using Windows page files with the IO Accelerator 64  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Performance and tuning  
Introduction to performance and tuning  
HP IO Accelerator devices provide high bandwidth and high IOPS and are specifically designed to achieve  
low latency.  
As IO Accelerator devices improve in IOPS and low latency, the device performance may be limited by  
operating system settings and BIOS configuration. To take advantage of the revolutionary performance of IO  
Accelerator devices, you might have to tune these settings.  
While IO Accelerator devices generally perform well out of the box, this section describes some of the  
common areas where tuning may help achieve optimal performance.  
Disabling DVFS  
DVFS is a power management technique that adjusts the CPU voltage and frequency to reduce power  
consumption by the CPU. These techniques help conserve power and reduce the heat generated by the CPU,  
but they adversely affect performance while the CPU transitions between low-power and high-performance  
states.  
These power-savings techniques are known to have a negative impact on I/O latency and maximum IOPS.  
When tuning for maximum performance, you might benefit from reducing or disabling DVSF completely,  
even though this might increase power consumption.  
DVFS, if available, should be configurable as part of your operating systems power management features as  
well as within your system BIOS interface. Within the operating system and BIOS, DVFS features are often  
found under the ACPI sections. Consult your computer documentation for details.  
Limiting APCI C-states  
Newer processors have the ability to go into lower power modes when they are not fully utilized. These idle  
states are known as ACPI C-states. The C0 state is the normal, full power, operating state. Higher C-states  
(C1, C2, C3, and so on) are lower power states.  
While ACPI C-states save on power, they are known to have a negative impact on I/O latency and maximum  
IOPS. With each higher C-state, typically more processor functions are limited to save power, and it takes  
time to restore the processor to the C0 state.  
These power savings techniques are known to have a negative impact on I/O latency and maximum IOPS.  
When tuning for maximum performance you might benefit from limiting the C-states or turning them off  
completely, even though this might increase power consumption.  
If your processor has ACPI C-states available, you can typically limit or disable them in the BIOS interface  
(sometimes referred to as a Setup Utility). APCI C-states might be part of the ACPI menu. For details, see your  
computer documentation.  
Performance and tuning 65  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Setting NUMA affinity  
Servers with a NUMA (Non-Uniform Memory Access) architecture require special installation instructions in  
order to maximize ioMemory device performance. These servers include the HP ProLiant DL580 and HP  
DL980 Servers.  
On servers with NUMA architecture, during system boot, the BIOS on some systems will not distribute PCIe  
slots evenly among the NUMA nodes. Each NUMA node contains multiple CPUs. This imbalanced  
distribution means that, during high workloads, half or more of the CPUs might remain idle while the rest are  
100% utilized. To prevent this imbalance, you must manually assign IO Accelerator devices equally among  
the available NUMA nodes.  
For information on setting NUMA affinity, see "NUMA configuration (on page 67)."  
Setting the interrupt handler affinity  
Device latency can be affected by placement of interrupts on NUMA systems. HP recommends placing  
interrupts for a given device on the same NUMA socket that the application is issuing I/O from. If the CPUs  
on this socket are overwhelmed with user application tasks, in some cases it might benefit performance to  
move the interrupts to a remote socket to help load balance the system.  
Many operating systems will attempt to dynamically place interrupts for you and generally make good  
decisions. Hand tuning interrupt placement is an advanced option that requires profiling of application  
performance on any given hardware. For information on how to pin interrupts for a given device to specific  
CPUs, see your operating system documentation.  
Performance and tuning 66  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
NUMA configuration  
Introduction to NUMA architecture  
Servers with NUMA (Non-Uniform Memory Access) architecture require special installation instructions in  
order to maximize IO Accelerator device performance. These servers include the HP DL580 and the HP  
DL980 server.  
On servers with NUMA architecture, during system boot, the BIOS on some systems will not distribute PCIe  
slots evenly among the NUMA nodes. Each NUMA node contains multiple CPUs. This imbalanced  
distribution means that, during high workloads, half or more of the CPUs will remain idle while the rest are  
100% utilized. To prevent this imbalance, you must manually assign IO Accelerator devices equally among  
the available NUMA nodes.  
Configuring the IO Accelerator devices for servers with NUMA architecture requires the use of the  
FIO_AFFINTIYparameter with the fio-configutility.  
FIO_AFFINITY parameter  
The FIO_AFFINITYparameter is a list of <affinity specification>triplets that specify the affinity  
settings of all adapters in the system. Each item in the triplet is separated by a comma, and each triplet set  
is separated by a semicolon.  
Syntax  
fio-config -p FIO_AFFINITY <affinity specification>[;<affinity  
specification>...]  
Where each <affinity specification>has the following syntax:  
<fct number>,[g|n]<group or node number>[,<hex mask>]  
If there is no gor ncharacter before the group or node number, then the number is assumed to be a group  
number.  
The hex mask is optional. If it is not present, the mask is assumed to be 0xffffffffffffffff. Also, the  
0xprefix is optional.  
If the hex mask is a node mask, then the mask is relative to the node, not the group to which the node belongs.  
Simple example  
fio-config -p FIO_AFFINITY 4,n1,0xf;5,n0;7,g1;9,g2,0xff0  
This command creates the following:  
Device  
fct4  
fct5  
fct7  
fct9  
Node/Group Processor Affinity  
node 1  
node 0  
group 1  
group 2  
Processors 0 to 3 in the node (mask 0xf)  
All processors in the node (no hex mask)  
All processors in the group (no hex mask)  
Processors 4 to 11 in the group (mask 0xff0)  
NUMA configuration 67  
Download from Www.Somanuals.com. All Manuals Search And Download.  
   
Advanced configuration example  
This example server has four NUMA nodes with eight hyper-threaded cores per node (16 logical processors  
per node, a total of 64 logical processors in the system). This system also uses the expansion configuration  
and has 11 PCIe expansion slots. During system boot, the system BIOS assigns PCIe slots 1-6 to NUMA node  
2 and PCIe slots 7-11 to NUMA node 0. NUMA nodes 1 and 3 have no assigned PCIe slots. This  
configuration creates a load balancing problem in the system when IO Accelerator devices are under heavy  
traffic. During these periods of high use, half of the processors in the system sit idle while the other half of the  
processors are 100% utilized, thus limiting the throughput of the IO Accelerator devices.  
To avoid this situation, you must manually configure the affinity of the IO Accelerator devices using the  
FIO_AFFINITYconfiguration parameter to distribute the work load across all NUMA nodes. This  
parameter overrides the default behavior of the IO Accelerator driver. For more information about the  
FIO_AFFINITYconfiguration parameter, refer to the syntax explanation below.  
Syntax:  
The following is an example of how to configure 10 HP IO Accelerator ioDrive Duo devices (each with two  
IO Accelerator devices) in a HP DL580 G7 system manually as described in the preceding paragraphs. Slot  
1 is a Generation 1 PCIe slot, so it is not compatible with an ioDrive Duo device. Therefore you can fill slots  
2-11 with ioDrive Duo devices. Because each ioDrive Duo device has two IO Accelerator devices, each  
ioDrive Duo devices has two device numbers (one for each IO Accelerator device). Each slot has two device  
numbers.  
The following tables list the default BIOS NUMA node assignments.  
BIOS assigned PCIe slots  
NUMA node  
FCT device numbers  
Processor Affinity  
7-11  
8,9,13,14,18,19,23,24,28,29  
None  
All processors in the node  
None  
0
None  
2-6  
1
2
135,136,140,141,145,146,150,151, All processors in the node  
155,156  
None  
None  
None  
3
Assigned  
PCIe slots  
FCT device numbers  
Processor Affinity  
NUMA node  
7-9  
8,9,13,14,18,19  
All processors in the node (no hex mask)  
All processors in the node (no hex mask)  
All processors in the node (no hex mask)  
All processors in the node (no hex mask)  
0
1
2
3
10-11  
2-3  
23,24,28,29  
135,136,140,141  
145,146,150,151,155,156  
4-6  
In this example, the BIOS creates a load imbalance by assigning the cards to only two NUMA nodes in the  
system. To balance the work load, enter the following settings:  
To configure the IO Accelerator driver with these override settings, run fio-configwith the following  
string:  
fio-config -p FIO_AFFINITY  
8,n0;9,n0;13,n0;14,n0;18,n0;19,n0;23,n1;24,n1;28,n1;29,n1;135,n2;136,n2;140,  
n2;141,n2;145,n3;146,n3;150,n3;151,n3;155,n3;156,n3  
No <hex mask>was stipulated for any of the devices in this example (making each <affinity  
specification>a couplet rather than a triplet). Therefore, each device is shared among all of the  
NUMA configuration 68  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
processors in the specified nodes. To fine tune the configuration, you can stipulate the hex mask in each  
<affinity specification>and specify the processors for each device.  
Checking the log for errors  
If you enter a configuration that is not valid, the settings will be disabled and an error will be display in the  
system logs.  
Example:  
fio-config -p FIO_AFFINITY 5,g0,0xf;6,0xf  
In this example, the affinity for device fct6is set incorrectly, because there is no group or node number  
before the mask. The following errors appear in the system log:  
2011-09-09T12:22:15.176086800Z - ERROR - FusionEventDriver - FIO_AFFINITY:  
Invalid group or node number  
2011-09-09T12:22:15.176086800Z - ERROR - FusionEventDriver - Invalid  
FIO_AFFINITY parameter syntax at character 13: "5,g0,0xf;6,0x". Manual  
affinity settings are disabled!  
NUMA configuration 69  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Resources  
Subscription service  
HP recommends that you register your product at the Subscriber’s Choice for Business website  
After registering, you will receive e-mail notification of product enhancements, new driver versions, firmware  
updates, and other product resources.  
For more information  
For additional information, see the following HP websites:  
HP BladeSystem technical resources (http://www.hp.com/go/bladesystem/documentation) (white  
papers and support documents)  
HP BladeSystem components  
Resources 70  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Support and other resources  
Before you contact HP  
Be sure to have the following information available before you call HP:  
Active Health System log (HP ProLiant Gen8 or later products)  
Download and have available an Active Health System log for 3 days before the failure was detected.  
For more information, see the HP iLO 4 User Guide or HP Intelligent Provisioning User Guide on the HP  
Onboard Administrator SHOW ALL report (for HP BladeSystem products only)  
For more information on obtaining the Onboard Administrator SHOW ALL report, see the HP website  
Technical support registration number (if applicable)  
Product serial number  
Product model name and number  
Product identification number  
Applicable error messages  
Add-on boards or hardware  
Third-party hardware or software  
Operating system type and revision level  
HP contact information  
For United States and worldwide contact information, see the Contact HP website  
In the United States:  
To contact HP by phone, call 1-800-334-5144. For continuous quality improvement, calls may be  
recorded or monitored.  
If you have purchased a Care Pack (service upgrade), see the Support & Drivers website  
(http://www8.hp.com/us/en/support-drivers.html). If the problem cannot be resolved at the website,  
call 1-800-633-3600. For more information about Care Packs, see the HP website  
Customer Self Repair  
HP products are designed with many Customer Self Repair (CSR) parts to minimize repair time and allow for  
greater flexibility in performing defective parts replacement. If during the diagnosis period HP (or HP service  
Support and other resources 71  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
providers or service partners) identifies that the repair can be accomplished by the use of a CSR part, HP will  
ship that part directly to you for replacement. There are two categories of CSR parts:  
Mandatory—Parts for which customer self repair is mandatory. If you request HP to replace these parts,  
you will be charged for the travel and labor costs of this service.  
Optional—Parts for which customer self repair is optional. These parts are also designed for customer  
self repair. If, however, you require that HP replace them for you, there may or may not be additional  
charges, depending on the type of warranty service designated for your product.  
NOTE: Some HP parts are not designed for customer self repair. In order to satisfy the customer warranty,  
HP requires that an authorized service provider replace the part. These parts are identified as "No" in the  
Illustrated Parts Catalog.  
Based on availability and where geography permits, CSR parts will be shipped for next business day  
delivery. Same day or four-hour delivery may be offered at an additional charge where geography permits.  
If assistance is required, you can call the HP Technical Support Center and a technician will help you over the  
telephone. HP specifies in the materials shipped with a replacement CSR part whether a defective part must  
be returned to HP. In cases where it is required to return the defective part to HP, you must ship the defective  
part back to HP within a defined period of time, normally five (5) business days. The defective part must be  
returned with the associated documentation in the provided shipping material. Failure to return the defective  
part may result in HP billing you for the replacement. With a customer self repair, HP will pay all shipping  
and part return costs and determine the courier/carrier to be used.  
For more information about HP's Customer Self Repair program, contact your local service provider. For the  
North American program, refer to the HP website (http://www.hp.com/go/selfrepair).  
Réparation par le client (CSR)  
Les produits HP comportent de nombreuses pièces CSR (Customer Self Repair = réparation par le client) afin  
de minimiser les délais de réparation et faciliter le remplacement des pièces défectueuses. Si pendant la  
période de diagnostic, HP (ou ses partenaires ou mainteneurs agréés) détermine que la réparation peut être  
effectuée à l'aide d'une pièce CSR, HP vous l'envoie directement. Il existe deux catégories de pièces CSR:  
Obligatoire - Pièces pour lesquelles la réparation par le client est obligatoire. Si vous demandez à HP de  
remplacer ces pièces, les coûts de déplacement et main d'œuvre du service vous seront facturés.  
Facultatif - Pièces pour lesquelles la réparation par le client est facultative. Ces pièces sont également  
conçues pour permettre au client d'effectuer lui-même la réparation. Toutefois, si vous demandez à HP de  
remplacer ces pièces, l'intervention peut ou non vous être facturée, selon le type de garantie applicable à  
votre produit.  
REMARQUE: Certaines pièces HP ne sont pas conçues pour permettre au client d'effectuer lui-même la  
réparation. Pour que la garantie puisse s'appliquer, HP exige que le remplacement de la pièce soit effectué  
par un Mainteneur Agréé. Ces pièces sont identifiées par la mention "Non" dans le Catalogue illustré.  
Les pièces CSR sont livrées le jour ouvré suivant, dans la limite des stocks disponibles et selon votre situation  
géographique. Si votre situation géographique le permet et que vous demandez une livraison le jour même  
ou dans les 4 heures, celle-ci vous sera facturée. Pour bénéficier d'une assistance téléphonique, appelez le  
Centre d'assistance technique HP. Dans les documents envoyés avec la pièce de rechange CSR, HP précise  
s'il est nécessaire de lui retourner la pièce défectueuse. Si c'est le cas, vous devez le faire dans le délai  
indiqué, généralement cinq (5) jours ouvrés. La pièce et sa documentation doivent être retournées dans  
l'emballage fourni. Si vous ne retournez pas la pièce défectueuse, HP se réserve le droit de vous facturer les  
coûts de remplacement. Dans le cas d'une pièce CSR, HP supporte l'ensemble des frais d'expédition et de  
retour, et détermine la société de courses ou le transporteur à utiliser.  
Support and other resources 72  
Download from Www.Somanuals.com. All Manuals Search And Download.  
Pour plus d'informations sur le programme CSR de HP, contactez votre Mainteneur Agrée local. Pour plus  
d'informations sur ce programme en Amérique du Nord, consultez le site Web HP  
Riparazione da parte del cliente  
Per abbreviare i tempi di riparazione e garantire una maggiore flessibilità nella sostituzione di parti  
difettose, i prodotti HP sono realizzati con numerosi componenti che possono essere riparati direttamente  
dal cliente (CSR, Customer Self Repair). Se in fase di diagnostica HP (o un centro di servizi o di assistenza  
HP) identifica il guasto come riparabile mediante un ricambio CSR, HP lo spedirà direttamente al cliente per  
la sostituzione. Vi sono due categorie di parti CSR:  
Obbligatorie – Parti che devono essere necessariamente riparate dal cliente. Se il cliente ne affida la  
riparazione ad HP, deve sostenere le spese di spedizione e di manodopera per il servizio.  
Opzionali – Parti la cui riparazione da parte del cliente è facoltativa. Si tratta comunque di componenti  
progettati per questo scopo. Se tuttavia il cliente ne richiede la sostituzione ad HP, potrebbe dover sostenere  
spese addizionali a seconda del tipo di garanzia previsto per il prodotto.  
NOTA: alcuni componenti HP non sono progettati per la riparazione da parte del cliente. Per rispettare la  
garanzia, HP richiede che queste parti siano sostituite da un centro di assistenza autorizzato. Tali parti sono  
identificate da un "No" nel Catalogo illustrato dei componenti.  
In base alla disponibilità e alla località geografica, le parti CSR vengono spedite con consegna entro il  
giorno lavorativo seguente. La consegna nel giorno stesso o entro quattro ore è offerta con un supplemento  
di costo solo in alcune zone. In caso di necessità si può richiedere l'assistenza telefonica di un addetto del  
centro di supporto tecnico HP. Nel materiale fornito con una parte di ricambio CSR, HP specifica se il cliente  
deve restituire dei componenti. Qualora sia richiesta la resa ad HP del componente difettoso, lo si deve  
spedire ad HP entro un determinato periodo di tempo, generalmente cinque (5) giorni lavorativi. Il  
componente difettoso deve essere restituito con la documentazione associata nell'imballo di spedizione  
fornito. La mancata restituzione del componente può comportare la fatturazione del ricambio da parte di HP.  
Nel caso di riparazione da parte del cliente, HP sostiene tutte le spese di spedizione e resa e sceglie il  
corriere/vettore da utilizzare.  
Per ulteriori informazioni sul programma CSR di HP contattare il centro di assistenza di zona. Per il  
programma in Nord America fare riferimento al sito Web HP (http://www.hp.com/go/selfrepair).  
Customer Self Repair  
HP Produkte enthalten viele CSR-Teile (Customer Self Repair), um Reparaturzeiten zu minimieren und höhere  
Flexibilität beim Austausch defekter Bauteile zu ermöglichen. Wenn HP (oder ein HP Servicepartner) bei der  
Diagnose feststellt, dass das Produkt mithilfe eines CSR-Teils repariert werden kann, sendet Ihnen HP dieses  
Bauteil zum Austausch direkt zu. CSR-Teile werden in zwei Kategorien unterteilt:  
Zwingend – Teile, für die das Customer Self Repair-Verfahren zwingend vorgegeben ist. Wenn Sie den  
Austausch dieser Teile von HP vornehmen lassen, werden Ihnen die Anfahrt- und Arbeitskosten für diesen  
Service berechnet.  
Optional – Teile, für die das Customer Self Repair-Verfahren optional ist. Diese Teile sind auch für Customer  
Self Repair ausgelegt. Wenn Sie jedoch den Austausch dieser Teile von HP vornehmen lassen möchten,  
können bei diesem Service je nach den für Ihr Produkt vorgesehenen Garantiebedingungen zusätzliche  
Kosten anfallen.  
Support and other resources 73  
Download from Www.Somanuals.com. All Manuals Search And Download.  
HINWEIS: Einige Teile sind nicht für Customer Self Repair ausgelegt. Um den Garantieanspruch des  
Kunden zu erfüllen, muss das Teil von einem HP Servicepartner ersetzt werden. Im illustrierten Teilekatalog  
sind diese Teile mit „No“ bzw. „Nein“ gekennzeichnet.  
CSR-Teile werden abhängig von der Verfügbarkeit und vom Lieferziel am folgenden Geschäftstag geliefert.  
Für bestimmte Standorte ist eine Lieferung am selben Tag oder innerhalb von vier Stunden gegen einen  
Aufpreis verfügbar. Wenn Sie Hilfe benötigen, können Sie das HP technische Support Center anrufen und  
sich von einem Mitarbeiter per Telefon helfen lassen. Den Materialien, die mit einem CSR-Ersatzteil geliefert  
werden, können Sie entnehmen, ob das defekte Teil an HP zurückgeschickt werden muss. Wenn es  
erforderlich ist, das defekte Teil an HP zurückzuschicken, müssen Sie dies innerhalb eines vorgegebenen  
Zeitraums tun, in der Regel innerhalb von fünf (5) Geschäftstagen. Das defekte Teil muss mit der zugehörigen  
Dokumentation in der Verpackung zurückgeschickt werden, die im Lieferumfang enthalten ist. Wenn Sie das  
defekte Teil nicht zurückschicken, kann HP Ihnen das Ersatzteil in Rechnung stellen. Im Falle von Customer  
Self Repair kommt HP für alle Kosten für die Lieferung und Rücksendung auf und bestimmt den  
Kurier-/Frachtdienst.  
Weitere Informationen über das HP Customer Self Repair Programm erhalten Sie von Ihrem Servicepartner  
vor Ort. Informationen über das CSR-Programm in Nordamerika finden Sie auf der HP Website unter  
Reparaciones del propio cliente  
Los productos de HP incluyen muchos componentes que el propio usuario puede reemplazar (Customer Self  
Repair, CSR) para minimizar el tiempo de reparación y ofrecer una mayor flexibilidad a la hora de realizar  
sustituciones de componentes defectuosos. Si, durante la fase de diagnóstico, HP (o los proveedores o socios  
de servicio de HP) identifica que una reparación puede llevarse a cabo mediante el uso de un componente  
CSR, HP le enviará dicho componente directamente para que realice su sustitución. Los componentes CSR se  
clasifican en dos categorías:  
Obligatorio: componentes para los que la reparación por parte del usuario es obligatoria. Si solicita a  
HP que realice la sustitución de estos componentes, tendrá que hacerse cargo de los gastos de  
desplazamiento y de mano de obra de dicho servicio.  
Opcional: componentes para los que la reparación por parte del usuario es opcional. Estos  
componentes también están diseñados para que puedan ser reparados por el usuario. Sin embargo, si  
precisa que HP realice su sustitución, puede o no conllevar costes adicionales, dependiendo del tipo de  
servicio de garantía correspondiente al producto.  
NOTA: Algunos componentes no están diseñados para que puedan ser reparados por el usuario. Para que  
el usuario haga valer su garantía, HP pone como condición que un proveedor de servicios autorizado  
realice la sustitución de estos componentes. Dichos componentes se identifican con la palabra "No" en el  
catálogo ilustrado de componentes.  
Según la disponibilidad y la situación geográfica, los componentes CSR se enviarán para que lleguen a su  
destino al siguiente día laborable. Si la situación geográfica lo permite, se puede solicitar la entrega en el  
mismo día o en cuatro horas con un coste adicional. Si precisa asistencia técnica, puede llamar al Centro de  
asistencia técnica de HP y recibirá ayuda telefónica por parte de un técnico. Con el envío de materiales  
para la sustitución de componentes CSR, HP especificará si los componentes defectuosos deberán  
devolverse a HP. En aquellos casos en los que sea necesario devolver algún componente a HP, deberá  
hacerlo en el periodo de tiempo especificado, normalmente cinco días laborables. Los componentes  
defectuosos deberán devolverse con toda la documentación relacionada y con el embalaje de envío. Si no  
enviara el componente defectuoso requerido, HP podrá cobrarle por el de sustitución. En el caso de todas  
Support and other resources 74  
Download from Www.Somanuals.com. All Manuals Search And Download.  
sustituciones que lleve a cabo el cliente, HP se hará cargo de todos los gastos de envío y devolución de  
componentes y escogerá la empresa de transporte que se utilice para dicho servicio.  
Para obtener más información acerca del programa de Reparaciones del propio cliente de HP, póngase en  
contacto con su proveedor de servicios local. Si está interesado en el programa para Norteamérica, visite  
la página web de HP siguiente (http://www.hp.com/go/selfrepair).  
Customer Self Repair  
Veel onderdelen in HP producten zijn door de klant zelf te repareren, waardoor de reparatieduur tot een  
minimum beperkt kan blijven en de flexibiliteit in het vervangen van defecte onderdelen groter is. Deze  
onderdelen worden CSR-onderdelen (Customer Self Repair) genoemd. Als HP (of een HP Service Partner) bij  
de diagnose vaststelt dat de reparatie kan worden uitgevoerd met een CSR-onderdeel, verzendt HP dat  
onderdeel rechtstreeks naar u, zodat u het defecte onderdeel daarmee kunt vervangen. Er zijn twee  
categorieën CSR-onderdelen:  
Verplicht: Onderdelen waarvoor reparatie door de klant verplicht is. Als u HP verzoekt deze onderdelen  
voor u te vervangen, worden u voor deze service reiskosten en arbeidsloon in rekening gebracht.  
Optioneel: Onderdelen waarvoor reparatie door de klant optioneel is. Ook deze onderdelen zijn ontworpen  
voor reparatie door de klant. Als u echter HP verzoekt deze onderdelen voor u te vervangen, kunnen  
daarvoor extra kosten in rekening worden gebracht, afhankelijk van het type garantieservice voor het  
product.  
OPMERKING: Sommige HP onderdelen zijn niet ontwikkeld voor reparatie door de klant. In verband met  
de garantievoorwaarden moet het onderdeel door een geautoriseerde Service Partner worden vervangen.  
Deze onderdelen worden in de geïllustreerde onderdelencatalogus aangemerkt met "Nee".  
Afhankelijk van de leverbaarheid en de locatie worden CSR-onderdelen verzonden voor levering op de  
eerstvolgende werkdag. Levering op dezelfde dag of binnen vier uur kan tegen meerkosten worden  
aangeboden, indien dit mogelijk is gezien de locatie. Indien assistentie gewenst is, belt u een HP Service  
Partner om via de telefoon technische ondersteuning te ontvangen. HP vermeldt in de documentatie bij het  
vervangende CSR-onderdeel of het defecte onderdeel aan HP moet worden geretourneerd. Als het defecte  
onderdeel aan HP moet worden teruggezonden, moet u het defecte onderdeel binnen een bepaalde  
periode, gewoonlijk vijf (5) werkdagen, retourneren aan HP. Het defecte onderdeel moet met de  
bijbehorende documentatie worden geretourneerd in het meegeleverde verpakkingsmateriaal. Als u het  
defecte onderdeel niet terugzendt, kan HP u voor het vervangende onderdeel kosten in rekening brengen. Bij  
reparatie door de klant betaalt HP alle verzendkosten voor het vervangende en geretourneerde onderdeel en  
kiest HP zelf welke koerier/transportonderneming hiervoor wordt gebruikt.  
Neem contact op met een Service Partner voor meer informatie over het Customer Self Repair programma  
van HP. Informatie over Service Partners vindt u op de HP website (http://www.hp.com/go/selfrepair).  
Reparo feito pelo cliente  
Os produtos da HP são projetados com muitas peças para reparo feito pelo cliente (CSR) de modo a  
minimizar o tempo de reparo e permitir maior flexibilidade na substituição de peças com defeito. Se,  
durante o período de diagnóstico, a HP (ou fornecedores/parceiros de serviço da HP) concluir que o reparo  
pode ser efetuado pelo uso de uma peça CSR, a peça de reposição será enviada diretamente ao cliente.  
Existem duas categorias de peças CSR:  
Obrigatória – Peças cujo reparo feito pelo cliente é obrigatório. Se desejar que a HP substitua essas peças,  
serão cobradas as despesas de transporte e mão-de-obra do serviço.  
Support and other resources 75  
Download from Www.Somanuals.com. All Manuals Search And Download.  
Opcional – Peças cujo reparo feito pelo cliente é opcional. Essas peças também são projetadas para o  
reparo feito pelo cliente. No entanto, se desejar que a HP as substitua, pode haver ou não a cobrança de  
taxa adicional, dependendo do tipo de serviço de garantia destinado ao produto.  
OBSERVAÇÃO: Algumas peças da HP não são projetadas para o reparo feito pelo cliente. A fim de  
cumprir a garantia do cliente, a HP exige que um técnico autorizado substitua a peça. Essas peças estão  
identificadas com a marca "No" (Não), no catálogo de peças ilustrado.  
Conforme a disponibilidade e o local geográfico, as peças CSR serão enviadas no primeiro dia útil após o  
pedido. Onde as condições geográficas permitirem, a entrega no mesmo dia ou em quatro horas pode ser  
feita mediante uma taxa adicional. Se precisar de auxílio, entre em contato com o Centro de suporte técnico  
da HP para que um técnico o ajude por telefone. A HP especifica nos materiais fornecidos com a peça CSR  
de reposição se a peça com defeito deve ser devolvida à HP. Nos casos em que isso for necessário, é  
preciso enviar a peça com defeito à HP dentro do período determinado, normalmente cinco (5) dias úteis.  
A peça com defeito deve ser enviada com a documentação correspondente no material de transporte  
fornecido. Caso não o faça, a HP poderá cobrar a reposição. Para as peças de reparo feito pelo cliente, a  
HP paga todas as despesas de transporte e de devolução da peça e determina a transportadora/serviço  
postal a ser utilizado.  
Para obter mais informações sobre o programa de reparo feito pelo cliente da HP, entre em contato com o  
fornecedor de serviços local. Para o programa norte-americano, visite o site da HP  
Support and other resources 76  
Download from Www.Somanuals.com. All Manuals Search And Download.  
Support and other resources 77  
Download from Www.Somanuals.com. All Manuals Search And Download.  
Support and other resources 78  
Download from Www.Somanuals.com. All Manuals Search And Download.  
Regulatory information  
Safety and regulatory compliance  
For safety, environmental, and regulatory information, see Safety and Compliance Information for Server,  
Storage, Power, Networking, and Rack Products, available at the HP website  
Turkey RoHS material content declaration  
Ukraine RoHS material content declaration  
Warranty information  
HP ProLiant and X86 Servers and Options (http://www.hp.com/support/ProLiantServers-Warranties)  
Regulatory information 79  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Acronyms and abbreviations  
ACPI  
Advanced Configuration and Power Interface  
DVFS  
dynamic voltage and frequency scaling  
FAT  
file allocation table  
IOPS  
input/output operations per second  
LEB  
Logical Erase Block  
LFM  
linear feet per minute  
MIB  
management information base  
MSI  
Message Signaled Interrupt  
NAND  
Not AND  
NTFS  
NT File System  
NUMA  
Non-Uniform Memory Architecture  
PCIe  
peripheral component interconnect express  
Acronyms and abbreviations 80  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
SMI-S  
Storage Management Initiative Specification  
SSD  
solid-state drive  
VM  
Virtual Machine  
VSL  
virtual storage layer  
Acronyms and abbreviations 81  
Download from Www.Somanuals.com. All Manuals Search And Download.  
Documentation feedback  
HP is committed to providing documentation that meets your needs. To help us improve the documentation,  
send any errors, suggestions, or comments to Documentation Feedback (mailto:[email protected]).  
Include the document title and part number, version number, or the URL when submitting your feedback.  
Documentation feedback 82  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
Index  
A
F
about this guide 6  
FIO_AFFINITY parameters 44, 67  
fio-attach utility 41  
fio-beacon utility 42  
Adding a file system 28  
authorized reseller 71  
automated logging 35  
fio-bugreport utility 42  
fio-config options 44  
fio-config utility 43  
fio-detach utility 45  
fio-format utility 46  
fio-pci-check utility 47  
fio-status utility 47  
fio-sure-erase utility 50  
fio-trim-config utility 51  
fio-update-iodrive utility 52  
firmware version 23  
B
battery replacement notice 79  
before you contact HP 71  
C
checking for outdated firmware 26  
configuring IO Accelerator paging support 60  
configuring RAID 29  
firmware, upgrading 19, 20  
flashback substitution events 55  
For more information 70  
formatting a file system 28  
contact information 71  
contacting HP 71  
contents summary 6  
controlling Trim aggressiveness 58  
creating a patch install log 34  
creating an installation log 34  
creating an uninstall log 34  
CSR (customer self repair) 71  
customer self repair (CSR) 71  
G
guaranteeing minimum committable memory 63  
H
health metrics 54  
D
health monitoring techniques 54  
HP Subscriber's Choice for Business 70  
HP technical support 71  
HP, contacting 71  
defragmentation 26  
disabling auto attach 37  
disabling DVFS 65  
disabling paging support 61  
documentation feedback 82  
I
installing software 13  
introduction 7  
E
enabling auto attach 38  
enabling paging support 61  
Enabling PCIe power 27  
enabling PCIe power override 38  
enabling Trim 58  
L
limiting APCI C-states 65  
error messages 69  
European Union notice 79  
M
maintenance tools 37  
Index 83  
Download from Www.Somanuals.com. All Manuals Search And Download.  
 
manual installation on Windows Server 21  
monitoring IO Accelerator health 54  
supported hardware 9  
system drive paging file configuration 63  
system log errors 69  
N
T
naming 27  
NAND flash and component failure 54  
non-paged memory pool 61  
NUMA architecture 67  
Taiwan battery recycling notice 79  
technical support 71  
telephone numbers 71  
Trim configurations 59  
Trim support 57  
NUMA configuration 67, 68  
Trim, introduction 57  
Trim, platforms 57  
troubleshooting 35  
O
operating environment 9  
outdated firmware 26  
overview 7  
U
uninstalling a previously-installed drive 12  
uninstalling the driver 37  
unmanaged shutdown issues 38  
upgrade procedures 23  
upgrading driver software 24, 25  
upgrading firmware 19, 20, 23  
using installation logs 33  
using test mode registry values 30  
using the IO Accelerator as a swap 29  
using Trim 57  
P
page files, setting up 62  
PCIe power override, enabling 38  
PCIe power, enabling 27  
performance and tuning 65  
performance and tuning, introduction 65  
performance attributes 8  
performing a multi-disk configuration 28  
phone numbers 71  
utilities 41  
utilities reference 41  
R
RAID configuration 29  
V
RAM consumption for page files 60  
regulatory compliance identification numbers 79  
regulatory compliance notices 79  
required information 71  
verifying page file operation 64  
Virtual Memory performance 64  
resources 70, 71  
W
Windows Installer logging options 33  
Windows page file management 62  
Windows page files 60  
Windows page files, introduction 60  
Windows Server 2003 22  
Windows Server 2008 23  
Windows Server 2012 23  
S
series number 79  
setting interrupt handler affinity 66  
setting NUMA affinity 66  
setting up SNMP 29  
setup wizard, using 14  
silent installation 26  
SNMP MIB support 31  
SNMP, Windows details 29  
software installation 13  
software RAID and health monitoring 55  
starting Trim 58  
Windows Server environments 12  
stopping Trim 58  
support 71  
supported firmware revisions 9  
Index 84  
Download from Www.Somanuals.com. All Manuals Search And Download.  

Graco Baby Playpen ISPP119AE User Manual
Grizzly Scale T10007 User Manual
Grundig Coffeemaker KM 4060 User Manual
Hatteras Hammocks Indoor Furnishings Tri Beam User Manual
HeadsetUSA Portable Speaker PAS 400 User Manual
HP Hewlett Packard Network Card 2240 User Manual
HP Hewlett Packard Network Router B User Manual
HP Hewlett Packard Switch 2520G User Manual
IBM Computer Accessories Version 52 User Manual
Icom Marine Radio IM402A User Manual