Posts tagged ·

7.1

·...

AIX 7.1 Installation Tips

Comments Off

This document contains the latest tips for successful installation of
AIX 7.1, and will be updated as new tips become available.


Technology Levels, Service Packs, APARs and PTFs mentioned in this document,
when available, can be obtained from the following web site.



http://www.ibm.com/eserver/support/fixes/fixcentral/main/pseries/aix


The AIX installation CD-ROMs and the level of AIX pre-installed on new
systems may not contain the latest fixes available at the time you install
the system, and may contain errors. Some these fixes may be critical to
the proper operation of your system. We recommend that you update to the
latest service pack level, which can be obtained from

http://www.ibm.com/eserver/support/fixes/fixcentral/main/pseries/aix.

When updating to a new Technology Level, it is a good practice to first
update the bos.rte.install update in a separate installation
session.

With certain combinations of updates, the update process may have to be run a
second time in order to apply all updates in a package. Checking the output of
the ‘oslevel -r’ and ‘oslevel -s’ commands for the expected values
after an update is recommended. Until the update is run a second time, the
output of the oslevel command may not indicate that the package is fully
installed.

The compare_report command, which is documented in the AIX
Commands Reference, can be used to determine which available updates are
newer than those installed on your system.

Any library or executable that is updated by an interim fix (emgr) or service
update, which is in use by an active process, will not be reflected in that
process until it is restarted. In addition, any process that is using a
library and does a dlopen() of the same library after the library has been
updated, could experience inconsistencies.


 

 



Free Space Requirements for devices.common.IBM.sni

The devices.common.IBM.sni product requires that 256MB of available, free
partitions in the rootvg volume group. If this space has not been previously
allocated for the sni product, a test will be performed at installation time
to make sure that this space is available, and if it is not, the installation
will fail. If installation is successful, a test will be performed when the
lpar is restarted after installation to determine if the sni hardware is
available, and if it is, a 256MB logical volume will be allocated as
/var/adm/sni.


7100-01 Installation

When applying the 7100-01 Technology Level with Service Pack 1 included,
you will have to run ‘smitty update_all’ a second time to update
bos.aso and mcr.rte. Until this is done, the ‘oslevel’ command
will not indicate the correct level.


7100-01 Service Pack 1

The 7100-01-01 Service Pack is considered highly recommended. 7100-01-01
is included in the 7100-01 Technology Level available on

Fix Central
, but systems installed prior to the availability of Service
Pack 1 should install 7100-01-01. To determine if the 7100-01-01 Service Pack
is installed, run the command:

    oslevel -s

The output should indicate “7100-01-01-1141“.


7100-01 Update When Using Cluster Aware AIX

Due to enhancements in Cluster Aware AIX (CAA) in the 6100-07 and 7100-01
Technology Levels, CAA must be stopped prior to the update. Attempting to
update to 6100-07 or 7100-01 with an active cluster will fail. Note that
PowerHA SystemMirror 7.1 exploits CAA. This procedure must be used when
performing one of the following with a CAA cluster deployed.

  • Updating from AIX 6.1 6100-06 to 6100-07
  • Updating from AIX 7.1 7100-00 to 7100-01
  • Migrating from AIX 6.1 6100-06 to AIX 7.1 7100-01

Update Procedure for PowerHA SystemMirror 7.1

  1. Backup the cluster configuration using the PowerHA SystemMirror
    snapshot facility.
  2. Stop the PowerHA Cluster Services through SMIT. Stop cluster
    services on all nodes in the cluster.
  3. Remove the CAA cluster using the command
    rmcluster -n ClusterName'.
  4. Update the AIX Technology Level and reboot the cluster nodes.
  5. Verify and Synchronize the CAA cluster configuration from
    PowerHA SystemMirror.
  6. Restart the PowerHA SystemMirror 7.1 Cluster.

Update Procedure for Cluster Aware AIX

  1. Save the CAA cluster configuration.
  2. Remove the CAA cluster using the command
    rmcluster -n ClusterName'.
  3. Update the AIX Technology Level and reboot the cluster nodes.
  4. Redeploy the CAA cluster using the mkcluster command.

Refer to the AIX_TL_update_information_for_PowerHA_7.1.pdf document at
the following location for additional information.


ftp://public.dhe.ibm.com/aix/efixes/iv08163/


7100-00 Service Pack 1

Service Pack 1 is considered a highly recommended update for customers
installing
AIX 7.1. To determine if you already have Service Pack 1 installed, use
the command:

    oslevel -s

The output should include “7100-00-01″. Service Pack 1 can be obtained from
the

Fix Central
web site.


Comments Off

SAS 9.2 on IBM AIX 7.1: Tuning guideline

Comments Off

SAS 9.2 on IBM AIX 7.1:

Tuning guidelines

 

This paper provides a starting point for performance optimization from a system-wide perspective to create an enhanced environment for SAS 9.2 on IBM® POWER® processor-based servers that run IBM AIX® 7.1.

 

§ Technology levels:

Use the latest version of AIX 7 Version 7.1 with the latest maintenance level and fix pack. Refer to the following website: ibm.com/support/fixcentral

At publication of this paper, TL00 with fix pack is considered to be the latest technology level.

 

§ IBM Hardware Managements Console (HMC):

Keep the HMC and microcode up-to-date.

You can search update bulletins for HMC and microcode at:

www14.software.ibm.com/webapp/set2/subscriptions/pqvcmjd

 

§ VMM:

AIX 7.1 default VMM parameters are mostly tuned for optimal SAS performance, except for one parameter. If needed, you can change the other values by using the AIX vmo command. It is suggested that you start with the following parameter value:

nokilluid=10

minfree=use greater of 960 or (128 * number of logical processors)

maxfree=(minfree + j2_maxPageReadAhead * number of logical processors)

Note: The following have been designated as restricted values in AIX 7 and should not be changed unless asked by AIX support. These values are already in-line with what SAS recommends.

 

  • maxperm%
  • maxclient%
  • strict_maxclient
  • strict_maxperm

 

§ I/O:

Tune I/O at the filesystem layer by using the AIX ioo command to enable Enhanced Journaled File System (JFS2) to perform efficient caching. It is suggested that you start with the following parameter values:

j2_dynamicBufferPreallocation=256

j2_nBufferPerPagerDevice=2048 j2_maxPageReadAhead=1024 j2_minPageReadAhead=16

j2_nPagesPerWriteBehindCluster=64

 

For higher workloads, it is recommended to increase the value of j2_maxPageReadAhead upto 2048.

 

§ Network:

Tune network parameters by using the no command. If SAS applications (such as SPDS, SPDE and/or SAS/CONNECT) are being used, set the following parameter value:

tcp_nodelayack=1

 

 

§ Maximum user process:

If the maximum number of processes for a single user exceeds 2000, increase the value of maxuproc to prevent SAS processes from abnormal shutdown or delay.

Increase the maxuproc setting by using the AIX smit or chdev command, for example: chdev –l sys0 –a maxuproc=<new value>

 

 

§ User limits:

Increase user-process resource limits for SAS users and database instances as appropriate (for example, unlimited or some tuned value for all resources (Refer to the Note).

In the /etc/security/limits file, set -1 for all resources.

Note: The default AIX user-process resource limits might be too low for SAS power users or large enterprise-class deployments of SAS. When SAS processes end because of attempts to exceed these resource limits, the system administrator typically sets all user-process resource limits to unlimited (a numeric value of -1) for users who run SAS. The problem with this approach is that the increased multithreading and scalability support in newer versions of SAS, coupled with an unlimited setting for user-process resource limits, allows other users to potentially exhaust system resources such as processor, memory I/O, and paging space. Before increasing user-process resource limits, such as memory, to very high values, carefully consider the potential consequences.

 

 

§ Paging space:

Configure the paging space to include at least the following suggestions:

Place paging spaces on dedicated disks to eliminate I/O contention.

 

Use multiple paging spaces that are spread across multiple disks.

 

Make the primary paging space hd6 a little bigger than the secondary paging spaces.

 

Ensure that the paging space is sufficient to support the number of concurrent SAS processes (because the number of SAS processes can be dynamic, depending on application workload).

 

§ Volume groups (VGs):

Use the AIX Scalable or Big volume group.

The scalable VG implementation provides configuration flexibility with respect to the number of physical volumes (PVs) and logical volumes (LVs) that a given instance of the new VG type can accommodate. The configuration options allow any scalable VG to contain 32, 64, 128, 256, 512, 768, or 1024 disks and 256, 512, 1024, 2048, or 4096 LVs. You do not need to configure the maximum values of 1024 PVs and 4096 LVs when creating the VG to account for potential future growth. You can increase the initial settings later, as required.

 

§ Disk layout:

Minimize disk contention between SAS temporary space and data spaces.

Avoid disk contention by placing SAS temporary-space file systems and SAS data file systems on physically separate disks.

 

Use multiple storage-server controllers to further separate and isolate the I/O traffic between SAS temporary and data spaces.

 

Use multiple mount points for SAS file systems. Place the operating system, SAS, user, SAS temporary space and SAS data file systems on separate physical disks.

 

Consider creating multiple SAS WORK areas that can be used by groups of SAS users.

 

Create separate JFS2 log files on separate physical disks for each SAS file system.

 

Spread the I/O workload across many physical disk spindles rather than across fewer, larger-capacity disks. Determine the sizing, based on the quantity of disks rather than disk capacity. Do not wrap logical unit numbers (LUNs) around the same spindle sets.

 

Do not share disk spindles with a RDBMS.

 

 

§ Release-behind mechanism for JFS2:

This feature allows the file system to release the file pages from file system buffer cache as soon as an application has read or written the file pages.

 

 

This feature helps when the SAS application performs a great deal of sequential reads or writes and most often, once accessed, these file pages are not accessed again in the near future. This feature can be configured on a file system basis. When using the mount command, enable release-behind by specifying one of the three flags below:

Release-behind sequential read flag (-rbr),

 

Release-behind sequential write flag (-rbw),

 

Release-behind sequential read and write flag (-rbrw).

 

§ Host bus adapters (HBAs):

Use an adequate number of HBAs from storage to the host server to provide the required application bandwidth.

Consider high-performance storage channels, such as Fibre Channel technology instead of slower mediums.

 

If possible, use dynamic multipathing to spread the I/O load across multiple adapters.

 

§ Redundant Array of Independent Disks (RAID):

Implement storage system RAID striping across multiple physical disks.

Use RAID10 or RAID5, depending on the level of redundancy and total capacity instead of the usable capacity that is needed for each file system type.

 

Use Logical Volume Manager (LVM) striping instead of concatenation.

 

 

§ LVM striping:

When choosing the disk stripe or segment size, or array stripe size, note that AIX file systems are aligned on a 16 KB boundary.

A strip is the size of data to be written to each physical disk in the array. A stripe is the size of the full write across all the physical disks in the array. For example: strip size x number of disks = stripe size

 

The AIX LVM stripe size that you can select from the smit lv create panel is actually the single strip size (not stripe). In other words, it is the size of data to be written to each of the array disks; it is not the full stripe size across all the physical disks. Consider using an LVM stripe size of 64 K or 128 K. Stripe sizes of 256 K or 512 K has shown better I/O performance in the case of SAS 9.2 workloads.

 

Synchronize SAS BUFSIZE with the storage-system stripe size and the AIX LVM stripe size (if using LVM striping) and VMM read-ahead increments.

 

Synchronizing I/O sizes streamlines I/O processing and reduces the number of I/O requests to the storage subsystem.

 

 

§ Resources:

For detailed tuning information, refer to the following publications:

Best Practices for Configuring your IO Subsystem for SAS9 Applications http://support.sas.com/rnd/papers/sgf07/sgf2007-iosubsystem.pdf

How to Maintain Happy SAS Users http://support.sas.com/resources/papers/happyIT.pdf

 

AIX 6 best practices for SAS on IBM POWER6

http://www.sas.com

 

Comments Off