| Blogs | Classifieds | Downloads | FlashChat | Gallery | Googlemap | Invite Friends | Links | Projects | Reviews | Wiki |
| |||||||||
Welcome to the pSeries Tech Forums,
our free peer-based support site for administrators, engineers and architects working with IBM pSeries servers and software. You are currently viewing our site as a guest which gives you limited access to view most discussions, articles, tutorials and access our other free features. By joining our community you will be able to collaborate with administrators, engineers and architects charged with designing, delivering or maintaining IBM pSeries server environments. Founded by a recognized IBM pSeries consultant and IBM Redbook author, pSeries Tech Forums was developed with the single mission of bringing IBM pSeries professionals together into a single self-help community. Registration is fast, simple and absolutely free to all IT professionals with responsibility for or interest in IBM pSeries servers. We invite you to join our community today! If you have any problems with the registration process or your account login, please contact contact support. |
| Our Sponsors | |
| | |
| Want to advertise? | |
![]() |
| | LinkBack | Thread Tools |
|
#1
| ||||
| ||||
Hi All, Wondering if I can get some direction as to what to do next. I have a filesystem that is producing 100% disk busy and only writing 4Mb/Sec max. TPS is 330 peak with a size of 12K average. The data has it's own disk and the jfslog is on a seperate hdisk. The data is temporary files for Sonic ESB/MQ Application I've tried tuning memory and jfs2 options via vmo and ioo. But not had a major impact. Also tried disk pacing. Would DIO/CIO help or would it be better if I just put the filesystem on SAN storage? Cheers Ran. |
|
#2
| ||||
| ||||
>> Would DIO/CIO help Depends on the size of the writes / reads .. CIO is only really good with block sizes of 1MB and upwards .. http://www-03.ibm.com/servers/aix/wh...b_perf_aix.pdf Run filemon to see what is going on and what file/LV your bottleneck is and post some output .. Spreading the LV over more spindles is going probably going to be the quick fix .. unless you have contention on the same file / inode .. (you can use splat to analyse this also) - and sync_release_ilock may also be an option. http://publib.boulder.ibm.com/infoce...mds5/splat.htm you can tune ioo if you know what types of writes / reads you are doing i.e. sequential reads and sequential write and random writes can be tuned a little (random reads cannot) Rgds Mark Taylor |
|
#4
| ||||
| ||||
Hi Mark, I'm going to try putting the filesystem on part of the SAN we have. Failing this I might go for your suggestion about the ram disk. I've tried doing some tuning with some of the parameters but it does not make much of a difference. I think that's all the performance I'm going to get out of the one drive. Spreading the load over several spindles sounds the way to go, I think this should give us more tps. Many Thanks Ran. |
![]() |
| Bookmarks |
| Currently Active Users Viewing This Thread: 1 (0 members and 1 guests) | |
| Thread Tools | |
| |