Sponsored Content
Operating Systems AIX Lots of page faults and free memory Post 302380778 by flpgdt on Wednesday 16th of December 2009 06:55:00 AM
Old 12-16-2009
Hi guys,

Thanks a lot for the replies and sorry for the late response!

So yeah, I do agree that the setting as poor right now. Appart from so many page faults, the isn't a sensible impact in performance. Actually as zaxxon pointed out, there isn't a whole lot of I/O happening here.

As for the tuning recommendations, they seem right, in fact they are a lot more in accordance to what this white paper says:

http://www.ibm.com/developerworks/forums/servlet/JiveServlet/download/747-3

(page 14)

But to be honest I can't really touch that. I'm more of application/dev level, so I'm trying to figure out what I could tell from my pov what could be done.
For now, I can only report what I've found to be 'normal' according to the current settings, and forward the white book as a guide line.
Besides, these flags seem to have been changed already. The guys doing it had they reason and knowledge for that.

Nevertheless, I'll monitor any changes they make in the server and let you guys know as it gets sorted.

Still, I wish I could tell more precisely, i.e. 1+1=2, why I'm seeing so much page faults.
While I can see that 20% maxperm, would drive my LRU nut trying to free more space in a memory intense environment, I cant really grasp how it causes so many page faults...

A guy from the IBM forum following the same tread said that I probably wont be able to figure that out without deep knowledge of what's running. That's somewhat... unsaddling.. I mean, would that mean that to fine tune my memory manager, I have to be a specialist in everything that runs in a server plus the server itself? I'd prefer to believe in a JVM-like world, where I can monitor and test GC's statistics and figure out my settings without having to know a lot of what or how does the program does...

cheers,

f.

EDIT - removed grumpiness Smilie

Last edited by flpgdt; 12-16-2009 at 09:57 AM..
 

10 More Discussions You Might Find Interesting

1. HP-UX

Copy on Write page faults

Hello Please can you tell me how to access COPY ON WRITE page faults in HP -UNIX. I found the structure in /usr/include/sys/vmmeter with the structure name vmmeter. Please tell me the function to fill the values to this structure and also the arguments for function.:: (5 Replies)
Discussion started by: manjunath
5 Replies

2. HP-UX

Intransient blocking page faults

Hi, Will anybody tell me what is this 'intransient blocking page faults' in HP-UX, it is in the structure _pst_vminfo in the header file /ust/include/sys/pstat/vm_pstat_body.h? (4 Replies)
Discussion started by: sushaga
4 Replies

3. UNIX for Advanced & Expert Users

Shared memory shortage but lots of unused memory

I am running HP-UX B.11.11. I'm increasing a parameter for a database engine so that it uses more memory to buffer the disk drive (to speed up performance). I have over 5GB of memory not being used. But when I try to start the DB with the increased buffer parameter I get told. "Not... (1 Reply)
Discussion started by: cjcamaro
1 Replies

4. AIX

High Page Faults

Sorry my poor english In 570 pseries nmon shows excessive page faults, ascents of something more than 30000 Page faults. System: AIX 5.2 ML5 Processor Type: PowerPC_POWER5 Number Of Processors: 2 Processor Clock Speed: 1656 MHz CPU Type: 64-bit Kernel Type: 64-bit Memory Size: 2816 MB ... (1 Reply)
Discussion started by: daviguez
1 Replies

5. Solaris

how to get the more memory free space (see memory free column)

Hi all, Could please let me know how to get the more memory free space (not added the RAM) in local zone. -bash-3.00# vmstat 2 5 kthr memory page disk faults cpu r b w swap free re mf pi po fr de sr s0 s1 s1 s1 in sy cs us sy... (3 Replies)
Discussion started by: murthy76
3 Replies

6. Programming

How to deal with lots of data in memory in order not to run out of memory

Hi, I'm trying to learn how to manage memory when I have to deal with lots of data. Basically I'm indexing a huge file (5GB, but it can be bigger), by creating tables that holds offset <-> startOfSomeData information. Currently I'm mapping the whole file at once (yep!) but of course the... (1 Reply)
Discussion started by: emitrax
1 Replies

7. AIX

Lots of page faults on AIX mySQL lpar

Hi, OS = AIX 5.3 Large number of page faults recently start to occure on AIX 5.3 lpar with mysql database installed. I need help in setting AIX OS parameter to solve the paging problem and some guidance on interpreting my stats t Code: # vmstat... (5 Replies)
Discussion started by: crosys
5 Replies

8. AIX

AIX 7.1 high page faults

hi guys i hope you can help me with this situation. i have 2 lpar with aix 7.1 and oracle 11gr2 in grid mode. when i start nmon to check the current system health i notice that page fault are over 3000/s. than i have opened a case with ibm and they say that the problem is not paging nor... (10 Replies)
Discussion started by: gullio23
10 Replies

9. Solaris

Page faults on OS

Hi guys, I have a zone on a M5000 server running solaris 10. The zone has an SAP application running on it and facing some performance issues. As part of the troubleshooting, I've been recommended to look for any paging on the OS. Please advise how to look for the paging. I've been looking at... (4 Replies)
Discussion started by: frum
4 Replies

10. AIX

High Paging when lots of free memory AIX 5.3

I am new to AIX, I have few AIX 5.3 servers and I could see there are significant difference in paging space utilization on servers even though they are running same applications below server is working fine which shows 2-5 % paging usage throuh out the day cpu_scale_memp = 8... (12 Replies)
Discussion started by: bibish
12 Replies
PIDSTAT(1)                                                      Linux User's Manual                                                     PIDSTAT(1)

NAME
pidstat - Report statistics for Linux tasks. SYNOPSIS
pidstat [ -C comm ] [ -d ] [ -h ] [ -I ] [ -l ] [ -p { pid [,...] | SELF | ALL } ] [ -r ] [ -s ] [ -t ] [ -T { TASK | CHILD | ALL } ] [ -u ] [ -V ] [ -w ] [ interval [ count ] ] DESCRIPTION
The pidstat command is used for monitoring individual tasks currently being managed by the Linux kernel. It writes to standard output activities for every task selected with option -p or for every task managed by the Linux kernel if option -p ALL has been used. Not select- ing any tasks is equivalent to specifying -p ALL but only active tasks (tasks with non-zero statistics values) will appear in the report. The pidstat command can also be used for monitoring the child processes of selected tasks. Read about option -T below. The interval parameter specifies the amount of time in seconds between each report. A value of 0 (or no parameters at all) indicates that tasks statistics are to be reported for the time since system startup (boot). The count parameter can be specified in conjunction with the interval parameter if this one is not set to zero. The value of count determines the number of reports generated at interval seconds apart. If the interval parameter is specified without the count parameter, the pidstat command generates reports continuously. You can select information about specific task activities using flags. Not specifying any flags selects only CPU activity. OPTIONS
-C comm Display only tasks whose command name includes the string comm. This string can be a regular expression. -d Report I/O statistics (kernels 2.6.20 and later only). The following values are displayed: PID The identification number of the task being monitored. kB_rd/s Number of kilobytes the task has caused to be read from disk per second. kB_wr/s Number of kilobytes the task has caused, or shall cause to be written to disk per second. kB_ccwr/s Number of kilobytes whose writing to disk has been cancelled by the task. This may occur when the task truncates some dirty pagecache. In this case, some IO which another task has been accounted for will not be happening. Command The command name of the task. -h Display all activities horizontally on a single line. This is intended to make it easier to be parsed by other programs. -I In an SMP environment, indicate that tasks CPU usage (as displayed by option -u ) should be divided by the total number of proces- sors. -l Display the process command name and all its arguments. -p { pid [,...] | SELF | ALL } Select tasks (processes) for which statistics are to be reported. pid is the process identification number. The SELF keyword indi- cates that statistics are to be reported for the pidstat process itself, whereas the ALL keyword indicates that statistics are to be reported for all the tasks managed by the system. -r Report page faults and memory utilization. When reporting statistics for individual tasks, the following values are displayed: PID The identification number of the task being monitored. minflt/s Total number of minor faults the task has made per second, those which have not required loading a memory page from disk. majflt/s Total number of major faults the task has made per second, those which have required loading a memory page from disk. VSZ Virtual Size: The virtual memory usage of entire task in kilobytes. RSS Resident Set Size: The non-swapped physical memory used by the task in kilobytes. %MEM The tasks's currently used share of available physical memory. Command The command name of the task. When reporting global statistics for tasks and all their children, the following values are displayed: PID The identification number of the task which is being monitored together with its children. minflt-nr Total number of minor faults made by the task and all its children, and collected during the interval of time. majflt-nr Total number of major faults made by the task and all its children, and collected during the interval of time. Command The command name of the task which is being monitored together with its children. -s Report stack utilization. The following values are displayed: PID The identification number of the task being monitored. StkSize The amount of memory in kilobytes reserved for the task as stack, but not necessarily used. StkRef The amount of memory in kilobytes used as stack, referenced by the task. Command The command name of the task. -t Also display statistics for threads associated with selected tasks. This option adds the following values to the reports: TGID The identification number of the thread group leader. TID The identification number of the thread being monitored. -T { TASK | CHILD | ALL } This option specifies what has to be monitored by the pidstat command. The TASK keyword indicates that statistics are to be reported for individual tasks (this is the default option) whereas the CHILD keyword indicates that statistics are to be globally reported for the selected tasks and all their children. The ALL keyword indicates that statistics are to be reported for individual tasks and globally for the selected tasks and their children. Note: Global statistics for tasks and all their children are not available for all options of pidstat. Also these statistics are not necessarily relevant to current time interval: The statistics of a child process are collected only when it finishes or it is killed. -u Report CPU utilization. When reporting statistics for individual tasks, the following values are displayed: PID The identification number of the task being monitored. %usr Percentage of CPU used by the task while executing at the user level (application), with or without nice priority. Note that this field does NOT include time spent running a virtual processor. %system Percentage of CPU used by the task while executing at the system level (kernel). %guest Percentage of CPU spent by the task in virtual machine (running a virtual processor). %CPU Total percentage of CPU time used by the task. In an SMP environment, the task's CPU usage will be divided by the total num- ber of CPU's if option -I has been entered on the command line. CPU Processor number to which the task is attached. Command The command name of the task. When reporting global statistics for tasks and all their children, the following values are displayed: PID The identification number of the task which is being monitored together with its children. usr-ms Total number of milliseconds spent by the task and all its children while executing at the user level (application), with or without nice priority, and collected during the interval of time. Note that this field does NOT include time spent running a virtual processor. system-ms Total number of milliseconds spent by the task and all its children while executing at the system level (kernel), and col- lected during the interval of time. guest-ms Total number of milliseconds spent by the task and all its children in virtual machine (running a virtual processor). Command The command name of the task which is being monitored together with its children. -V Print version number then exit. -w Report task switching activity (kernels 2.6.23 and later only). The following values are displayed: PID The identification number of the task being monitored. cswch/s Total number of voluntary context switches the task made per second. A voluntary context switch occurs when a task blocks because it requires a resource that is unavailable. nvcswch/s Total number of non voluntary context switches the task made per second. A involuntary context switch takes place when a task executes for the duration of its time slice and then is forced to relinquish the processor. Command The command name of the task. ENVIRONMENT
The pidstat command takes into account the following environment variable: S_TIME_FORMAT If this variable exists and its value is ISO then the current locale will be ignored when printing the date in the report header. The pidstat command will use the ISO 8601 format (YYYY-MM-DD) instead. EXAMPLES
pidstat 2 5 Display five reports of CPU statistics for every active task in the system at two second intervals. pidstat -r -p 1643 2 5 Display five reports of page faults and memory statistics for PID 1643 at two second intervals. pidstat -C "fox|bird" -r -p ALL Display global page faults and memory statistics for all the processes whose command name includes the string "fox" or "bird". pidstat -T CHILD -r 2 5 Display five reports of page faults statistics at two second intervals for the child processes of all tasks in the system. Only child processes with non-zero statistics values are displayed. BUGS
/proc filesystem must be mounted for the pidstat command to work. FILES
/proc contains various files with system statistics. AUTHOR
Sebastien Godard (sysstat <at> orange.fr) SEE ALSO
sar(1), top(1), ps(1), mpstat(1), iostat(1), vmstat(8) http://pagesperso-orange.fr/sebastien.godard/ Linux AUGUST 2011 PIDSTAT(1)
All times are GMT -4. The time now is 05:56 PM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy