Sponsored Content
Full Discussion: AIX 6.1 memory tuning
Operating Systems AIX AIX 6.1 memory tuning Post 302887753 by bakunin on Monday 10th of February 2014 04:13:12 PM
Old 02-10-2014
Quote:
Originally Posted by dzodzo
vmstat -v shows that about 20% of memory pages are pinned (that would roughly correspond to those 13 GB for kernel). Does it mean that application doesn't use memory pinning (server is running Oracle+SAP)? svmon tells me that oracle and workprocesses use about 33 MB of pinned memory, perhaps it's the way they are designed. I'll have to check on some testing system what happens, if you continually increase memory for aplication, how it will affect OS behaviour.
hmmm....

First off: Oracle is indeed using "pinned memory", because "pinned memory" is normal memory, but not allowed to be swapped out. Oracle uses it for its "SGA" (system global area) on one hand and for shared memory on the other. If you are interested in the details of allocated shared memory i suggest you use the ipcs command to analyze which process owns which shared memory segment. I can warmly recommend the man page of ipcs, it is a phantastic read.

Second: yes, the kernel accumulates memory over time, but for a different reason: "file memory" is part of the memory accounted to the kernel too, because the kernel "owns" it, so to say. When the system starts and hasn't done anything it has no idea what to put into file cache, so it is initially empty. Over time it is filled and less important things get thrown out in favor of more important ones. The vmo parameters "lru_file_repage", "maxperm" and "minperm" steer the process and i suggest you read up on the vmo command (which sets these options) to understand the process better.

By the way, as Oracle has its own file caching mechanism built into the SGA it might be a wise idea to make the SGA bigger and diminish the AIX filecache accordingly. You might also consider changing the maxperm parameter to 97% instead of its current 90%, but this will probably not have a big effect if the shown values are typical for your machines load.

I hope this helps.

bakunin
 

10 More Discussions You Might Find Interesting

1. UNIX for Dummies Questions & Answers

Tuning Unix server memory - HPUX11i

Hello, I'm using a unix server (HP rp2450) which has : 2 Go RAM memory and 4 Go swap. Here is the result of vmstat -n command : $ vmstat -n VM memory page faults avm free re at pi po fr de sr in sy ... (5 Replies)
Discussion started by: thierryUX
5 Replies

2. AIX

TUNING: memory page sizes

This is post number 3 in a series of unanswered "TUNING" questions. :D With AIX 5.3 TL4, the page size can vary from the original "4k". They can now be "64k" and a couple other sizes. They also do not have to all be the same. Some can remain "4k" while others are "64k" which is what seems to... (2 Replies)
Discussion started by: kah00na
2 Replies

3. UNIX for Advanced & Expert Users

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... (1 Reply)
Discussion started by: theerthan
1 Replies

4. AIX

How to do Performance monitoring and tuning in AIX

How to do Performance monitoring and tuning in AIX. (2 Replies)
Discussion started by: AIXlearner
2 Replies

5. AIX

AIX Tuning For DB2

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

6. AIX

Tuning AIX for oracle

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

7. AIX

AIX file system tuning

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

8. AIX

New to AIX tcpip tuning

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

9. AIX

Network tuning parameters on 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

IBM AIX I/O Performance Tuning

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
ELVTUNE(8)						      System Manager's Manual							ELVTUNE(8)

NAME
elvtune - I/O elevator tuner SYNOPSIS
elvtune [ -r r_lat ] [ -w w_lat ] [ -b b_max ] /dev/blkdev1 [ /dev/blkdev2 ... ] elvtune -h elvtune -v DESCRIPTION
elvtune allows to tune the I/O elevator per blockdevice queue basis. The tuning can be safely done at runtime. Tuning the elevator means being able to change disk performance and interactiveness. In the output of elvtune the address of the queue tuned will be shown and it can be considered as a queue ID. For example multiple partitions in the same harddisk will share the same queue and so tuning one partition will be like tuning the whole HD. OPTIONS
-r r_lat set the max latency that the I/O scheduler will provide on each read. -w w_lat set the max latency that the I/O scheduler will provide on each write. -b b_max max coalescing factor allowed on writes when there are reads pending in the queue. -h help. -v version. NOTE
Actually the only fields tunable are those relative to the IO scheduler. It's not possible to select a one-way or two-way elevator yet. For logical blockdevices like LVM the tuning has to be done on the physical devices. Tuning the queue of the LVM logical device is useless. RETURN VALUE
0 on success and 1 on failure. HISTORY
Ioctls for tuning elevator behaviour were added in Linux 2.3.99-pre1. AUTHORS
Andrea Arcangeli <andrea@suse.de> SuSE Version 1.0 14 March 2000 ELVTUNE(8)
All times are GMT -4. The time now is 05:42 AM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy