Sponsored Content
Full Discussion: CIO/DIO and JFS2 read ahead
Operating Systems AIX CIO/DIO and JFS2 read ahead Post 302258257 by zaxxon on Friday 14th of November 2008 06:45:13 AM
Old 11-14-2008
As Shockneck said, CIO/DIO is said to be better for random reads/write afaik.

You could check with vmstat -v if there are still any blocked buffers counting up and tune them with the appropriate ioo parameters.
Also you can check about aio, if there is any usage of it or something close to it's limits with "iostat -A" and also the shift+a option in nmon, to see how many aioservers are being used currently.

Edit:
Ah I see we had a foregoing discussion already here
https://www.unix.com/unix-advanced-ex...evice-cio.html

Are you sure you need cio? You hadn't shown the "vmstat -v" back then still.

There is also the possibility to change attributes for FC adapter (like lg_term_dma, max_xfer_size and num_cmd_elems) and disks (like queue_depth or max_coalesce) with chdev and reading them out with lsattr. But before doing that, I would start to sort things out with AIO if appropriate and vmstat -v/iostat.

Last edited by zaxxon; 11-14-2008 at 08:23 AM.. Reason: Added something, and something more
 

8 More Discussions You Might Find Interesting

1. HP-UX

Read-ahead in HP-UX

One cool thing about unix is that it predicts disk blocks that you may need and tries to have them in core before you need them. Over the years, various unix vendors tried various algorithms to improve performance. HP has patented their latest algorithm... Multi-threaded Read Ahead Prediction... (0 Replies)
Discussion started by: Perderabo
0 Replies

2. AIX

JFS and JFS2

hi all, can sumbody give me a link which gives the basic layout of JFS, JFS2 and the veritas file system. and i also want to know about the data structures used in this filsystem thanx in advance (0 Replies)
Discussion started by: anwerreyaz
0 Replies

3. Shell Programming and Scripting

PERL look ahead and behind ...

I would like to search a router config file for "ip address $ip", once found, I want to grab the line just before that contains "interface $interfacetype" basically saying, 10.3.127.9 is assigned to "Loopback1" given the below as an example. interface Loopback1 ip address 10.3.127.9... (1 Reply)
Discussion started by: popeye
1 Replies

4. AIX

Jfs and jfs2

Hi all, Can anyone define the difference between jfs and jfs2 filesystem as well as usage of jfs log files.... Thanks.... (7 Replies)
Discussion started by: sumathi.k
7 Replies

5. Red Hat

How to enable AIO and DIO on rhel5 64bit?

Hi Friends, Please help me to understand, how to enable async disk IO and Direct disk IO in ext3 filesystem on rhel5. Regards, Arumon (0 Replies)
Discussion started by: arumon
0 Replies

6. AIX

Slow NFS when cio/dio enabled

Hi, I have a bit of a NFS problem on AiX 6.1 : When I set the mount to cio and dio - needed for a database app - Everything slows down. The following is copying 700mb, top one is a normal mount bottom one is a mount with the cio/dio option enabled : # ./a.sh Wed Jan 11 11:41:24 GMT 2012... (5 Replies)
Discussion started by: AJCG1976
5 Replies

7. AIX

AIX JFS2 content

Hello! I have a LPAR with 51GB RAM topas shows 21% of memory for client pages (~11Gb) svmon -G show 2906772 client pages in use (~11Gb) but then i try to investigate per process client memory usage svmon -P -O filtertype=client and summarize inuse column, i get only 347880 pages... (1 Reply)
Discussion started by: sys
1 Replies

8. UNIX for Beginners Questions & Answers

Two days ahead

Hi, I have a code that will show one day ahead, how to make it show two days ahead p-dev1-db-tst:/$ day=$(TZ=IST-24 date +%d) p-dev1-db-tst:/$ echo $day 17 p-dev1-db-tst:/$ Regards, Adam (5 Replies)
Discussion started by: answer
5 Replies
vmstat(1M)                                                System Administration Commands                                                vmstat(1M)

NAME
vmstat - report virtual memory statistics SYNOPSIS
vmstat [-cipqsS] [disks] [ interval [count]] DESCRIPTION
vmstat reports virtual memory statistics regarding kernel thread, virtual memory, disk, trap, and CPU activity. On MP (multi-processor) systems, vmstat averages the number of CPUs into the output. For per-processor statistics, see mpstat(1M). vmstat only supports statistics for certain devices. For more general system statistics, use sar(1), iostat(1M), or sar(1M). Without options, vmstat displays a one-line summary of the virtual memory activity since the system was booted. During execution of the kernel status command, the state of the system can change. If relevant, a state change message is included in the vmstat output, in one of the following forms: <<device added: sd0>> <<device removed: sd0>> <<processors added: 1, 3>> <<processors removed: 1, 3>> See System Administration Guide: Advanced Administration for device naming conventions for disks. OPTIONS
The following options are supported: -c Report cache flushing statistics. This option is obsolete, and no longer meaningful. This option might be removed in a future version of Solaris. -i Report the number of interrupts per device. count and interval does not apply to the -i option. -p Report paging activity in details. This option will display the following, respectively: epi Executable page-ins. epo Executable page-outs. epf Executable page-frees. api Anonymous page-ins. apo Anonymous page-outs. apf Anonymous page-frees. fpi File system page-ins. fpo File system page-outs. fpf File system page-frees. When executed in a zone and if the pools facility is active, all of the above only report actitivity on the processors in the pro- cessor set of the zone's pool. -q Suppress messages related to state changes. -s Display the total number of various system events since boot. count and interval does not apply to the -s option. -S Report on swapping rather than paging activity. This option will change two fields in vmstat's ``paging'' display: rather than the ``re'' and ``mf'' fields, vmstat will report ``si'' (swap-ins) and ``so'' (swap-outs). OPERANDS
The following operands are supported: count Specifies the number of times that the statistics are repeated. count does not apply to the -i and -s options. disks Specifies which disks are to be given priority in the output (only four disks fit on a line). Common disk names are id, sd, xd, or xy, followed by a number (for example, sd2, xd0, and so forth). interval Specifies the last number of seconds over which vmstat summarizes activity. This number of seconds repeats forever. inter- val does not apply to the -i and -s options. EXAMPLES
Example 1: Using vmstat The following command displays a summary of what the system is doing every five seconds. example% vmstat 5 kthr memory page disk faults cpu r b w swap free re mf pi p fr de sr s0 s1 s2 s3 in sy cs us sy id 0 0 0 11456 4120 1 41 19 1 3 0 2 0 4 0 0 48 112 130 4 14 82 0 0 1 10132 4280 0 4 44 0 0 0 0 0 23 0 0 211 230 144 3 35 62 0 0 1 10132 4616 0 0 20 0 0 0 0 0 19 0 0 150 172 146 3 33 64 0 0 1 10132 5292 0 0 9 0 0 0 0 0 21 0 0 165 105 130 1 21 78 1 1 1 10132 5496 0 0 5 0 0 0 0 0 23 0 0 183 92 134 1 20 79 1 0 1 10132 5564 0 0 25 0 0 0 0 0 18 0 0 131 231 116 4 34 62 1 0 1 10124 5412 0 0 37 0 0 0 0 0 22 0 0 166 179 118 1 33 67 1 0 1 10124 5236 0 0 24 0 0 0 0 0 14 0 0 109 243 113 4 56 39 ^C example% The fields of vmstat's display are kthr Report the number of kernel threads in each of the three following states: r the number of kernel threads in run queue b the number of blocked kernel threads that are waiting for resources I/O, paging, and so forth w the number of swapped out lightweight processes (LWPs) that are waiting for processing resources to finish. memory Report on usage of virtual and real memory. swap available swap space (Kbytes) free size of the free list (Kbytes) page Report information about page faults and paging activity. The information on each of the following activities is given in units per second. re page reclaims -- but see the -S option for how this field is modified. mf minor faults -- but see the -S option for how this field is modified. pi kilobytes paged in po kilobytes paged out fr kilobytes freed de anticipated short-term memory shortfall (Kbytes) sr pages scanned by clock algorithm When executed in a zone and if the pools facility is active, all of the above (except for "de") only report activity on the processors in the processor set of the zone's pool. disk Report the number of disk operations per second. There are slots for up to four disks, labeled with a single letter and number. The letter indicates the type of disk (s = SCSI, i = IPI, and so forth); the number is the logical unit number. faults Report the trap/interrupt rates (per second). in interrupts sy system calls cs CPU context switches When executed in a zone and if the pools facility is active, all of the above only report actitivity on the processors in the processor set of the zone's pool. cpu Give a breakdown of percentage usage of CPU time. On MP systems, this is an average across all processors. us user time sy system time id idle time When executed in a zone and if the pools facility is active, all of the above only report actitivity on the processors in the processor set of the zone's pool. ATTRIBUTES
See attributes(5) for descriptions of the following attributes: +-----------------------------+-----------------------------+ | ATTRIBUTE TYPE | ATTRIBUTE VALUE | +-----------------------------+-----------------------------+ |Availability |SUNWcsu | +-----------------------------+-----------------------------+ |Interface Stability |See below. | +-----------------------------+-----------------------------+ Invocation is evolving. Human readable output is unstable. SEE ALSO
sar(1), iostat(1M), mpstat(1M), sar(1M), attributes(5) System Administration Guide: Basic Administration System Administration Guide: Advanced Administration NOTES
The sum of CPU utilization might vary slightly from 100 because of rounding errors in the production of a percentage figure. The -c option (Report cache flushing statistics) is not supported in this release. SunOS 5.10 20 Dec 2004 vmstat(1M)
All times are GMT -4. The time now is 11:19 AM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy