Comments on: Effect of adaptive_flushing http://www.mysqlperformanceblog.com/2009/12/04/effect-of-adaptive_flushing/ Percona's MySQL & InnoDB performance and scalability blog Sat, 11 Feb 2012 16:45:54 +0000 hourly 1 http://wordpress.org/?v=3.2.1 By: aaa456 http://www.mysqlperformanceblog.com/2009/12/04/effect-of-adaptive_flushing/comment-page-1/#comment-703472 aaa456 Wed, 30 Dec 2009 03:19:55 +0000 http://www.mysqlperformanceblog.com/?p=1798#comment-703472 How about the result of XtraDB for the workload with settings ” innodb_adaptive_flushing=OFF innodb_adaptive_checkpoint=estimate innodb_use_purge_thread=1 ” ? “estimate” is based on the different strategy, and it is not so much eagerly to flush… How about the result of XtraDB for the workload with settings
”
innodb_adaptive_flushing=OFF
innodb_adaptive_checkpoint=estimate
innodb_use_purge_thread=1
”
?

“estimate” is based on the different strategy, and it is not so much eagerly to flush…

]]> By: Yasufumi http://www.mysqlperformanceblog.com/2009/12/04/effect-of-adaptive_flushing/comment-page-1/#comment-694066 Yasufumi Fri, 11 Dec 2009 01:32:55 +0000 http://www.mysqlperformanceblog.com/?p=1798#comment-694066 How about the result of XtraDB for the workload with settings " innodb_adaptive_flushing=OFF innodb_adaptive_checkpoint=estimate innodb_use_purge_thread=1 " ? "estimate" is based on the different strategy, and it is not so much eagerly to flush... How about the result of XtraDB for the workload with settings
”
innodb_adaptive_flushing=OFF
innodb_adaptive_checkpoint=estimate
innodb_use_purge_thread=1
”
?

“estimate” is based on the different strategy, and it is not so much eagerly to flush…

]]> By: Baron Schwartz http://www.mysqlperformanceblog.com/2009/12/04/effect-of-adaptive_flushing/comment-page-1/#comment-693476 Baron Schwartz Thu, 10 Dec 2009 05:48:43 +0000 http://www.mysqlperformanceblog.com/?p=1798#comment-693476 Patrick, your question merited another post: http://www.mysqlperformanceblog.com/2009/12/09/why-delayed-flushing-can-result-in-less-work/ Patrick, your question merited another post: http://www.mysqlperformanceblog.com/2009/12/09/why-delayed-flushing-can-result-in-less-work/

]]> By: Patrick Casey http://www.mysqlperformanceblog.com/2009/12/04/effect-of-adaptive_flushing/comment-page-1/#comment-693434 Patrick Casey Thu, 10 Dec 2009 03:20:00 +0000 http://www.mysqlperformanceblog.com/?p=1798#comment-693434 That's really interesting stuff; am I reading it correctly though that adaptive flush actually increased the IOOPS? Looking at the IO graphs, it looks like both the peak IO rate and average IO rate were higher with adaptive flush nabled (assuming I'm reading properly). Seems to imply that if recovery time isn't a major factor, you're better off (for this workload at least) running w/o that option enabled? That’s really interesting stuff; am I reading it correctly though that adaptive flush actually increased the IOOPS? Looking at the IO graphs, it looks like both the peak IO rate and average IO rate were higher with adaptive flush nabled (assuming I’m reading properly).

Seems to imply that if recovery time isn’t a major factor, you’re better off (for this workload at least) running w/o that option enabled?

]]> By: MLBR http://www.mysqlperformanceblog.com/2009/12/04/effect-of-adaptive_flushing/comment-page-1/#comment-691858 MLBR Mon, 07 Dec 2009 14:21:07 +0000 http://www.mysqlperformanceblog.com/?p=1798#comment-691858 Great post! Great post!

]]>