Date: Mon, 25 Jul 2005 18:00:21 +0200 From: Bohdan Horst <nexus@hoth.amu.edu.pl> To: Mark Kirkwood <markir@paradise.net.nz> Cc: Chris <chrcoluk@gmail.com>, freebsd-stable@freebsd.org Subject: Re: FreeBSD IO Performance (was Re: Quality of FreeBSD) Message-ID: <20050725160021.GB613@aristo> In-Reply-To: <42E4BBCD.3060406@paradise.net.nz> References: <1dbad31505072105401c06bee6@mail.gmail.com> <42E099A0.3080101@paradise.net.nz> <3aaaa3a050724144831d5cff0@mail.gmail.com> <42E4BBCD.3060406@paradise.net.nz>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Jul 25, 2005 at 10:15:41PM +1200, Mark Kirkwood wrote: > Hmm - looks like I missed that thread, never mind - repeatability of > findings is sound scientific principle :-) > > With respect to changing the default for vfs.read_max - makes sense to > me, but it would be interesting to know if anyone has a system that > performs *worse* with it set to 16. > > regards > > Mark > > Chris wrote: > >Its been mentioned before and most experience the same as you by > >setting it to 16 a dramatic improvement in the sequential read, I > >currently run all my 5.x servers like this with no issues as a result. > > I am curious if the default will ever be changed. on my notebook: IBM TP 600x 5.4-RELEASE-p4 ad0: 38154MB <IC25N040ATCS04-0/CA4OA71A> [77520/16/63] at ata0-master UDMA33 APM enable (but same effect with disable APM) AAC unfortunatelly not supported :( anything above vfs.read_max=1 produce very annoying little sound (i can hear disk heads) when playing movies.. my 2 cents -- Bohdan 'Nexus' Horst
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20050725160021.GB613>