Date: Thu, 26 Jan 2012 18:24:25 +0100 From: Willem Jan Withagen <wjw@digiware.nl> To: Peter Maloney <peter.maloney@brockmann-consult.de> Cc: freebsd-fs@freebsd.org Subject: Re: Question about ZFS with log and cache on SSD with GPT Message-ID: <4F218C49.9050905@digiware.nl> In-Reply-To: <4F1C3597.4040009@digiware.nl> References: <4F193D90.9020703@digiware.nl> <20120121162906.0000518c@unknown> <4F1B0177.8080909@digiware.nl> <20120121230616.00006267@unknown> <4F1BC493.10304@brockmann-consult.de> <4F1C3597.4040009@digiware.nl>
next in thread | previous in thread | raw e-mail | index | archive | help
On 2012-01-22 17:13, Willem Jan Withagen wrote: > On 22-1-2012 9:10, Peter Maloney wrote: >> Am 21.01.2012 23:06, schrieb Alexander Leidinger: >> Here is an example from our FreeBSD forum: >> http://forums.freebsd.org/showthread.php?t=19093 > > Thanx for this thread, there is a lot of usefull info there. > pithy thing is to blow 66Mb, but then again on 40 or 120 Gb SSDs it is > only marginal. (Guess it stems from the time that HDs where 5Mb :) ) > > I'm still not really shure that that is needed it the bios has nothing > to do with these disks, as in our case: SSDs are only used as caches > under ZFS. > > Especially the testing methods are useful. They are of course valid for > any type partinioning... > So getting things right on this level is the first required. > >>> I create the first partition at the usual 63 sectors offset from the >>> start of the disk (track 1) which is /unaligned/ with the SSD erase >>> block. The second partition is set to start at sector 21030912 >>> (10767826944 bytes) which is /aligned/ with the SSD erase block. I did have some testing time last night... So I removed the SSD's from the pool. En then first filled them up with atleas 40G of garbage to make shure the SSDs logic considered them full. Then I created the partitions without alignment gpart create -s GPT ada2 gpart add -s 512M -t freebsd-zfs ada2 gpart add -t freebsd-zfs ada2 (Not shure if this would TRIM the SSD?) On both partitions I got no more that 110Mb/s write speed, whatever I tried. (writing 512Mb or 10Gb of data) Then I recreated the partitions as suggested in the above article: gpart create -s GPT ada2 gpart add -b 129024 -s 512M -t freebsd-zfs ada2 gpart add -t freebsd-zfs ada2 Lo and behold: the write speed was 220Mb/s. So it does really, really really matter. And testing is relatively simple. I was not able to trash the zpool itself, so it still runds under ashift=9. And if that given akward SSD access, that is still around. But the lowest layer should now atleast perform. I'll get more testing time on the box next week when others are done testing. And then I'll rebuild the pool to get ashift=12. Regards, --WjW
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4F218C49.9050905>