Date: Mon, 3 Mar 2008 02:18:05 -1000 (HST) From: Jeff Roberson <jroberson@chesapeake.net> To: Julian Elischer <julian@elischer.org> Cc: src-committers@FreeBSD.org, cvs-all@FreeBSD.org, cvs-src@FreeBSD.org Subject: Re: cvs commit: src/sys/kern sched_ule.c Message-ID: <20080303021658.L920@desktop> In-Reply-To: <47CB81BB.1080009@elischer.org> References: <200803020821.m228L0Yw042389@repoman.freebsd.org> <20080301222513.Y920@desktop> <20080302105958.GF67687@server.vk2pj.dyndns.org> <47CB81BB.1080009@elischer.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, 2 Mar 2008, Julian Elischer wrote: > Peter Jeremy wrote: > >> >>> Kris has done some excellent benchmarking as usual. Here you can see the >>> improvement in postgres depending on various scheduler debug settings: >>> >>> http://people.freebsd.org/~kris/scaling/pgsql-16cpu.png >> >> The improvement is quite substantial. Congratulations Jeff. > > the drop off after 14 is intersting.. It's contention on various kernel locks, most likely the proc lock. Scheduling algorithms which limit concurrency at contention can improve throughput in these cases but it is to the detriment of workloads which do not suffer the same contention. The ultimate solution will be to improve proc/sleepq/signal locking for 8.0. > >> >
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20080303021658.L920>