Date: Thu, 21 Dec 2006 15:14:04 +1300 From: Mark Kirkwood <markir@paradise.net.nz> To: freebsd-stable@freebsd.org Cc: Pieter de Goeje <pieter@degoeje.nl> Subject: Re: Cached file read performance with 6.2-PRERELEASE Message-ID: <4589EDEC.2040504@paradise.net.nz> In-Reply-To: <4589A921.90002@paradise.net.nz> References: <45888C68.10305@paradise.net.nz> <200612200816.51043.joao@matik.com.br> <4589128F.9030404@paradise.net.nz> <200612201536.25497.pieter@degoeje.nl> <4589A921.90002@paradise.net.nz>
next in thread | previous in thread | raw e-mail | index | archive | help
This is a multi-part message in MIME format. --------------060008050409040508000700 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Mark Kirkwood wrote: > Pieter de Goeje wrote: >> It would be more interesting to see how random access to a (cached) >> file performs in Linux vs FreeBSD, which seems a more logical pattern >> for a database. >> > > Agreed, and good point, I'll knock up a simple program to do random > and/or sequential access of a file and see what we get! > Here's a (very) simple program that does block reads sequentially or randomly. It probably needs a little polishing, but seems to work ok for the size of files we are interested in: i.e < a few GB (see attached): Results: ======== Compiled with CFLAGS=-O2 -march=i686 Gentoo - 2.6.18-gentoo-r3: ----------------------- $ ./readtest /data0/dump/file 8192 0 random reads: 100000 elapsed: 1.2646 io rate 647805551 bytes/s $ ./readtest /data0/dump/file 8192 1 sequential reads: 100000 elapsed: 1.1267 io rate 727075854 bytes/s FreeBSD - 6.2-PRERELEASE #7: Mon Nov 27 19:32:33 NZDT 2006 : ------------------------------------------------------------ ./readtest /data0/dump/file 8192 0 random reads: 100000 elapsed: 4.3669 io rate 187594060 bytes/s $ ./readtest /data0/dump/file 8192 1 sequential reads: 100000 elapsed: 1.9679 io rate 416283642 bytes/s So looks like we get faster overall results than dd (I guess not needing to send output anywhere helps)...also we seem to be slower in the random case too :-(. I ran these programs several times, typical results shown. Cheers Mark --------------060008050409040508000700--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4589EDEC.2040504>