Date: Sun, 9 Jan 2000 16:47:36 +0800 From: Greg Lehey <grog@lemis.com> To: Brad Knowles <blk@skynet.be> Cc: Brad Chisholm <sasblc@unx.sas.com>, freebsd-current@FreeBSD.ORG Subject: RAID benchmarks (was: Adaptec AAC-364 RAID controller support) Message-ID: <20000109164735.B515@mojave.worldwide.lemis.com> In-Reply-To: <v04220801b49c19a4e4b4@[195.238.21.69]>; from blk@skynet.be on Fri, Jan 07, 2000 at 11:33:20PM %2B0100 References: <200001072210.AA10196@newlgn49.unx.sas.com> <v04220801b49c19a4e4b4@[195.238.21.69]>
next in thread | previous in thread | raw e-mail | index | archive | help
On Friday, 7 January 2000 at 23:33:20 +0100, Brad Knowles wrote: > At 5:10 PM -0500 2000/1/7, Brad Chisholm wrote: > >> The Adaptec card looks like it might provide a superior solution. What's >> best way to proceed when trying to get documentation support from vendors? > > Look long and hard at software solutions with vinum before you > wed yourself to a particular hardware RAID controller. At the very > least, I'd encourage you to extensively benchmark all the candidate > solutions in your environment, before you make a choice. > > You can see my own benchmarks that I've done so far at > <http://www.shub-internet.org/brad/FreeBSD/vinum.html>. I hate to say this, but the results are meaningless for the sequential case; the version of rawio you used started all parallel sequential transfers at the same place, so the results showed primarily how much cache each device had. I've updated rawio and the port, and if you (anybody) are really interested in sequential performance, you should run the tests again. My take on this is that sequential performance is only interesting in a few unusual cases, such as streaming video, and the performance isn't nearly as good as you think as soon as you introduce more than one client process. Greg -- Finger grog@lemis.com for PGP public key See complete headers for address and phone numbers To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20000109164735.B515>