Ramon van Alteren wrote: > kashani wrote: >> Ramon van Alteren wrote: >> >>>> did NTPL help you guys? >>>> >>> >>> Strangely it actually hurt our mysql-performance. Although mysql AB >>> recommends it. Haven't done any recent testing however. >> >> That's really odd. I've never seen Mysql do anything, but get much >> much faster with NTPL at least with our work loads. Web servers, >> mostly selects, PHP, 600-1000 connections to each of the db server. >> Load went from 1.5 to .3 on our dual proc boxes when we moved from >> 2.4 to 2.6 w/NTPL. I suspect it's the number of connections we have >> that caused most of the benefit in our case. > > Let's start with "which version ?" > > I'd have to look through all testing docs for the period but I > remember that we tested NTPL with 4.0 and found it hurt our performance. NTPL was un-supported by MySQL on 4.0, see [1], and support started not much more than one year ago (prior patching the sources was needed) > > Similar work-load: webservers, reading from slaves, writing only to > the replication-master, PHP > Maybe different sizing? We have a very large database spanning well > over 35Gb now, with some extremely large tables in it. Most dbservers > are IO-bound not CPU-bound. then switching thread implementation model would make no difference ;-)
[1] http://bugs.mysql.com/bug.php?id=19785 -- [email protected] mailing list
