tag:blogger.com,1999:blog-3043493300793589377.post2703325631920990989..comments2024-03-21T22:39:18.222-07:00Comments on Yoshinori Matsunobu's blog: Linux I/O scheduler queue size and MyISAM performanceYoshinori Matsunobuhttp://www.blogger.com/profile/14180479977952026421noreply@blogger.comBlogger13125tag:blogger.com,1999:blog-3043493300793589377.post-50134924837360434742010-09-15T08:38:14.942-07:002010-09-15T08:38:14.942-07:00wow this is great I also managed to increase the Q...wow this is great I also managed to increase the Queue size to 200,000 and the noop scheduler performance increased a lot, I did not know that I could configure the Queue size, thank you very much!Sildenafil Citratehttp://www.citratesildenafil.com/noreply@blogger.comtag:blogger.com,1999:blog-3043493300793589377.post-52293037203978356522010-04-29T14:04:35.400-07:002010-04-29T14:04:35.400-07:00I just increased the queue size to 100,000 and my ...I just increased the queue size to 100,000 and my WinXP VM started 15 seconds faster- from 60 to 45 seconds. Worth investigating. Thanks a lot!extrasaltyhttps://www.blogger.com/profile/11387184308544165947noreply@blogger.comtag:blogger.com,1999:blog-3043493300793589377.post-80307495697073845032010-03-06T05:23:06.737-08:002010-03-06T05:23:06.737-08:00How about arrays with more then 2 spindles? For in...How about arrays with more then 2 spindles? For instance my Data Array has 34 in a raid 10.<br /><br /><br />anyway <a href="http://www.adminkernel.com/linux/" rel="nofollow">Dont forget to click link.</a>Sumanthttps://www.blogger.com/profile/11824801392530480579noreply@blogger.comtag:blogger.com,1999:blog-3043493300793589377.post-70905519549320514702009-07-30T10:53:20.806-07:002009-07-30T10:53:20.806-07:00Yoshinori,
i wonder if the change of the parameter...Yoshinori,<br />i wonder if the change of the parameter "queue size" affects the performance of other types of querys (select or update for example)?<br />Sorry for my bad english...<br />tanks.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-3043493300793589377.post-60127097986301223242009-07-13T03:42:24.779-07:002009-07-13T03:42:24.779-07:00Not needed. The change is dynamically updated.Not needed. The change is dynamically updated.Yoshinori Matsunobuhttps://www.blogger.com/profile/14180479977952026421noreply@blogger.comtag:blogger.com,1999:blog-3043493300793589377.post-17407369912389758752009-07-13T03:33:48.447-07:002009-07-13T03:33:48.447-07:00is it necessary to to shutdown the oracle db befor...is it necessary to to shutdown the oracle db before changing the value? And is it neccessary to reboot the server after chanign the value?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-3043493300793589377.post-39698095238904266082009-06-24T16:29:16.956-07:002009-06-24T16:29:16.956-07:00How about arrays with more then 2 spindles? For in...How about arrays with more then 2 spindles? For instance my Data Array has 34 in a raid 10. I would think that the spindle count would have an impact on what is optimumAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-3043493300793589377.post-47564408276589105212009-05-08T18:42:00.000-07:002009-05-08T18:42:00.000-07:00So a trade off between latency and throughput when...So a trade off between latency and throughput when you know your load isn't latency sensitive. Nice tweak.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-3043493300793589377.post-91888742544229623012009-05-08T12:44:00.000-07:002009-05-08T12:44:00.000-07:00"Yoshinori Matsunobu’s item illuminating Linux I/O..."Yoshinori Matsunobu’s item illuminating Linux I/O scheduler queue size and MyISAM performance attracted some comment..."<br /><br /><A HREF="http://www.pythian.com/news/2393/log-buffer-145-a-carnival-of-the-vanities-for-dbas" REL="nofollow">Log Buffer #145</A>Log Bufferhttp://www.pythian.com/news/about-log-buffer/noreply@blogger.comtag:blogger.com,1999:blog-3043493300793589377.post-54434290794193427792009-04-30T16:03:00.000-07:002009-04-30T16:03:00.000-07:00Hi Mark,
I haven't done yet. In theory it should...Hi Mark, <br /><br />I haven't done yet. In theory it should be higher than 128 but lower than 100,000. For example, when setting 10,000, avgqu-sz should be less than 10,000, sorting less than 10,000 i/o requests. Then the rest i/o requests are not sorted so it is less efficient than setting 100,000.Yoshinori Matsunobuhttps://www.blogger.com/profile/14180479977952026421noreply@blogger.comtag:blogger.com,1999:blog-3043493300793589377.post-58740926708001414342009-04-30T10:30:00.000-07:002009-04-30T10:30:00.000-07:00This is very interesting. Do you have results for ...This is very interesting. Do you have results for a queue size larger than 128 but not as large as 100,000?Mark Callaghanhttps://www.blogger.com/profile/09590445221922043181noreply@blogger.comtag:blogger.com,1999:blog-3043493300793589377.post-75257473579334915932009-04-30T06:07:00.000-07:002009-04-30T06:07:00.000-07:00Hi Peter,
I published detailed test scripts. If ...Hi Peter, <br /><br />I published detailed test scripts. If you need more information please let me know.<br />I didn't have time to fully test on noop/anticipatory/cfq scheduler for this test, but as far as running a couple of hours I saw similar behavior (=increasing queue size improved throughput for this test) on noop scheduler.Yoshinori Matsunobuhttps://www.blogger.com/profile/14180479977952026421noreply@blogger.comtag:blogger.com,1999:blog-3043493300793589377.post-81585467377599099782009-04-29T23:20:00.000-07:002009-04-29T23:20:00.000-07:00Yoshinori,
I really think you're getting into som...Yoshinori,<br /><br />I really think you're getting into some side effect here. If you would publish scripts I would be very curious to redo the run to see if I can repeat the same behavior. <br /><br />It also would be interesting to see the IO data from the test. When you're inserting data you have potential of doing reads and writes. Reads are issued one by one anyway so they can't be affected by queue depth. It is possible scheduler algorithm causes side effect of having myisam files writes delayed.Anonymousnoreply@blogger.com