Date: Tue, 26 Jan 2010 17:07:43 +0100 From: Ivan Voras <ivoras@freebsd.org> To: =?UTF-8?Q?Dag=2DErling_Sm=C3=B8rgrav?= <des@des.no> Cc: freebsd-hackers@freebsd.org Subject: Re: A 7 vs 8 db benchmark Message-ID: <9bbcef731001260807x674cf22cl484c6d06c235406a@mail.gmail.com> In-Reply-To: <86ljfkj3k5.fsf@ds4.des.no> References: <hjdega$a6d$1@ger.gmane.org> <86ljfkj3k5.fsf@ds4.des.no>
next in thread | previous in thread | raw e-mail | index | archive | help
2010/1/26 Dag-Erling Sm=C3=B8rgrav <des@des.no>: > Ivan Voras <ivoras@freebsd.org> writes: >> This is a bit old, I forgot to post it earlier: >> >> http://suckit.blog.hu/2009/10/05/freebsd_8_is_it_worth_to_upgrade >> >> Some interesting graphs there. Jeff Roberson said the bogus CPU >> topology shouldn't influence end-performance much. He thinks the >> lockmgr rewrite was responsible for most of the big performance >> difference. > > Were the benchmarks run with or without the topology patch? =C2=A0If with= out, > could you ask the author to re-run them with the patch? With - the benchmark author created the patch.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?9bbcef731001260807x674cf22cl484c6d06c235406a>