04-04-2008
Tuning AIX IO
Hi
I am trying to investigate a disk performance issue, and we are not seem to be hitting the right direction in our analysis.
This is a FC disk running on USP1000 HDS system. The application is an IO intensive application, but our opinion is that it is not performing due to perceived disk issues.
Analysis:
iostat on the disk shows over various iterations, something like:
Disks: %tm_act Kbps tps Kb_read Kb_wrtn time
hdisk40 27 257.5 50.1 2551 12900 0:01:12
hdisk40 7.8 305.6 20.8 9088 9244 0:02:12
hdisk40 0.9 14 2.1 160 680 0:03:12
hdisk40 0.7 8.4 1.3 16 488 0:04:12
hdisk40 2.6 205 9.8 11012 1288 0:05:12
hdisk40 8.9 1055 50.1 57140 6160 0:06:12
hdisk40 6.1 1117.4 49.1 63432 3608 0:07:12
hdisk40 5.6 810.9 35.1 46488 2164 0:08:12
hdisk40 0.7 6.3 1 0 380 0:09:12
hdisk40 7.3 320 12.9 4988 14208 0:10:12
hdisk40 29.4 1751.5 83.3 19948 85128 0:11:12
To calculate the throughput here, is it OK to just take (Kb_read+Kb_wrtn)/1024 to get the throughput in MBPS?
Is iostat a reliable way to calculate throughput or do I have to necessarily use nstress?
Is there a problem with this disk?
Thanks
Theerthan
10 More Discussions You Might Find Interesting
1. Web Development
We have been tuning MySQL lately and I ran accoss two useful tools that you might be interested in:
mysqltuner.pl
tuning-primer.sh
Both of these scripts are quite useful for MySQL tuning. Here is some sample output of mysqltuner.pl
>> MySQLTuner 0.9.8 - Major Hayden... (3 Replies)
Discussion started by: Neo
3 Replies
2. AIX
How to do Performance monitoring and tuning in AIX. (2 Replies)
Discussion started by: AIXlearner
2 Replies
3. AIX
Dear friends. can anybody suggest me what to be considered in order to achieve maximum performance of AIX on which DB2 will be installed
Thanks is advance :) (1 Reply)
Discussion started by: Vit0_Corleone
1 Replies
4. AIX
Hi all,
From Googling, I found that the basics used for troubleshooting UNIX/AIX performance issues are commands like vmstat, iostat and sar. I believe these are generic commands regardless of what UNIX flavour is in used, only difference being is the format of the output.
In a real case... (2 Replies)
Discussion started by: newbie_01
2 Replies
5. AIX
Dears i want to have a clear view about this tuning parameters and what they related to FS or Oracle , and how to figure the percentage of them .
maxperm%
maxclient%
v_pinshm = 1
lgpg_regions = 0
lpgp_size = 0 (3 Replies)
Discussion started by: thecobra151
3 Replies
6. AIX
how do i determine the percentage of the following befor i install oracle 11g
maxperm%
maxclient%
v_pinshm
lgpg_regions
lgpg_size (3 Replies)
Discussion started by: thecobra151
3 Replies
7. AIX
For some reason, my AIX 5.2 box has become slow in accepting telnet requests from others boxes. Windows, times out the connection, whereas, Unix it will wait for the AIX to display the login. I connect and it respawns back and says connected, but then sits and wait for what seems forever to get the... (5 Replies)
Discussion started by: mrmurdock
5 Replies
8. AIX
Greetings,
i'm wondering if there is a way to determine minimum memory requirement for AIX kernel and OS functions? We use memdetails script from perfpmr package to see actual memory allocation, for example like this:
===========================================================================... (5 Replies)
Discussion started by: dzodzo
5 Replies
9. AIX
Hi,
we've a gigabit Ethernet adapter. And we wanted to improve the performance by tuning network parameters. so' as per IBM info center,
http://publib.boulder.ibm.com/infocenter/aix/v7r1/index.jsp?topic=%2Fcom.ibm.aix.prftungd%2Fdoc%2Fprftungd%2Fnetw_opt.htm
we tried changing the tuning... (2 Replies)
Discussion started by: System Admin 77
2 Replies
10. AIX
I have a IBM Power9 server coupled with a NVMe StorWize V7000 GEN3 storage, doing some benchmarks and noticing that single thread I/O (80% Read / 20% Write, common OLTP I/O profile) seems slow.
./xdisk -R0 -r80 -b 8k -M 1 -f /usr1/testing -t60 -OD -V
BS Proc AIO read% IO Flag IO/s ... (8 Replies)
Discussion started by: c3rb3rus
8 Replies
LEARN ABOUT HPUX
tunefs_hfs
tunefs(1M) tunefs(1M)
NAME
tunefs - tune up an existing HFS file system
SYNOPSIS
maxcontig] rotdelay] maxbpg] minfree]
advanced read-ahead] special-device
DESCRIPTION
The command is used to alter dynamic parameters that affect HFS file system layout policies. Parameters to be altered are specified by the
options and arguments provided on the command line as described below.
affects how the file system blocks are laid out on the disk. The default rotdelay value set by the and commands (see newfs(1M) and
mkfs(1M)) is 0 milliseconds, causing file system blocks to be written and read consecutively. In general, this should be the optimal tun-
ing, making the use of unnecessary.
Options
recognizes the following options and command-line arguments:
Set the maximum number of contiguous blocks that will be laid out
before forcing a rotational delay to maxcontig (see below). The default value is because most device drivers require
one interrupt per disk transfer. For device drivers that can chain several buffers together in a single transfer,
set maxcontig to the maximum chain length.
rotdelay is the expected time (in milliseconds) to service a transfer completion interrupt and initiate a new transfer on the
same disk. It is used to determine how much rotational spacing to place between successive blocks in a file.
maxbpg specifies the maximum number of blocks any single file can allocate out of a cylinder group before it is forced to
begin allocating blocks from another cylinder group. Typically this value is set to about one fourth of the total
blocks in a cylinder group. The intent is to prevent any single file from using up all the blocks in a single cylin-
der group, thus degrading access times for all files subsequently allocated in that cylinder group. The effect of
this limit is to cause large files to do long seeks more frequently than if they were allowed to allocate all the
blocks in a cylinder group before seeking elsewhere. For file systems with exclusively large files, this parameter
should be set higher.
minfree specifies the percentage of space that is not available to normal users; i.e., the minimum free space threshold. The
default value used is 10%. This value can be set to zero. If set to zero, throughput performance drops to as little
as one-third of the efficiency expected when the threshold is set at 10%. Note that if minfree is raised above the
current usage level, users cannot allocate files until enough files have been deleted to meet the new threshold
requirement.
Advanced read-ahead
specifies whether the file system should use an advanced predictive read-ahead algorithm. The implementation
requires more system resources in exchange for an advanced access pattern recognition. Patterns include forward
sequential, backward sequential, forward strided, and backward strided. This value can be set to zero (disable) or
one (enable). By default, a file system will have advanced read-ahead enabled when created.
(visual) Display current values
contained in the primary super-block to standard output.
(all) Modify redundant super-blocks as well as the primary super-block
as stipulated by the configuration options and arguments.
special-device is the name of the file system to be tuned. It is either a block or character special file if the file system is not
mounted, or a block special file if the file system is mounted.
WARNINGS
Root file system tuning is normally done during initial system software installation. Tuning the root file system after installation has
little useful effect because so many files have already been written.
AUTHOR
was developed by the University of California, Berkeley.
SEE ALSO
dumpfs(1M), mkfs(1M), newfs(1M).
tunefs(1M)