Date: Sat, 28 Jun 2014 12:08:39 +0200 From: Palle Girgensohn <girgen@pingpong.net> To: Konstantin Belousov <kostikbel@gmail.com> Cc: "freebsd-current@freebsd.org" <freebsd-current@freebsd.org>, "performance@freebsd.org" <performance@freebsd.org> Subject: Re: PostgreSQL performance on FreeBSD Message-ID: <35090A62-2DB8-493C-A5ED-ADB1BC193640@pingpong.net> In-Reply-To: <20140627163407.GX93733@kib.kiev.ua> References: <20140627125613.GT93733@kib.kiev.ua> <201406271057.53599.jhb@freebsd.org> <20140627163407.GX93733@kib.kiev.ua>
next in thread | previous in thread | raw e-mail | index | archive | help
27 jun 2014 kl. 18:34 skrev Konstantin Belousov <kostikbel@gmail.com>: > On Fri, Jun 27, 2014 at 10:57:53AM -0400, John Baldwin wrote: >> On Friday, June 27, 2014 8:56:13 am Konstantin Belousov wrote: >>> Hi, >>> I did some measurements and hacks to see about the performance and >>> scalability of PostgreSQL 9.3 on FreeBSD, sponsored by The FreeBSD >>> Foundation. >>>=20 >>> The results are described in https://kib.kiev.ua/kib/pgsql_perf.pdf. >>> The uncommitted patches, referenced in the article, are available as >>> https://kib.kiev.ua/kib/pig1.patch.txt >>> https://kib.kiev.ua/kib/patch-2 >>=20 >> Did you run the same benchmark on the same hardware with any other OS's t= o=20 >> compare results? >=20 > No. >=20 > FWIW, before the failing after the 30 clients is corrected, I do not > think it is much interesting to do such comparision. This is great work! Does anybody know how far back in FreeBSD versions using posix semaphore ins= tead of sysv would make a difference? It seems we need a rather current ver= sion? 8.x did not support it at all, at some point at lest, and in 9 it was b= uggy. I could add he patch-2 to the port, but I reckon it needs a conditiona= l based on FreeBSD version? The clang bug should go upstreams, right? I have seen similar curves, presented by Greg Smith (PostgreSQL hacker) wher= e he concluded that there is no point in running more than 50 concurrent con= nections. This was for Linux. In your measures, the knee is at 30. That's sa= id, FreeBSD could and should do better, but probably there is a limit where t= here will be a knee in the graph and performance will drop. It should be mor= e than 30, though, as you rightly commented. Do you any ideas to pursue this further apart from complicated rewrites like= DragonFly? Palle=
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?35090A62-2DB8-493C-A5ED-ADB1BC193640>