Date: Mon, 22 Mar 1999 09:26:28 +1030 From: Greg Lehey <grog@lemis.com> To: Tom <tom@sdf.com>, Nick Hilliard <nick@iol.ie> Cc: freebsd-scsi@FreeBSD.ORG Subject: Re: dpt raid-5 performance Message-ID: <19990322092628.V429@lemis.com> In-Reply-To: <Pine.BSF.4.05.9903211255260.24283-100000@misery.sdf.com>; from Tom on Sun, Mar 21, 1999 at 12:58:46PM -0800 References: <199903211417.OAA28733@beckett.earlsfort.iol.ie> <Pine.BSF.4.05.9903211255260.24283-100000@misery.sdf.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sunday, 21 March 1999 at 12:58:46 -0800, Tom wrote: > > On Sun, 21 Mar 1999, Nick Hilliard wrote: > >>> I haven't yet replied to Nick's message because I wanted to check >>> something here first, and I've been too busy so far. But I'll come >>> back with some comparisons. >> >> I'm going to run some benchmarks over the next few days and see what they >> throw up. >> >> My instinct was that 512K was a "good" interleave size in some sense of the >> word, mainly because of the fact that it would cause so many fewer disk io >> ops in most circumstances -- in fact, all circumstances except where you're >> doing piles of tiny io ops. The bonnie results seem to shatter this >> illusion. > > Uhh... no. Large stripe sizes for good for lots of parallel > processes. We're still out on that one. > Bonnie is only a single process, Bonnie's random seek test runs three parallel processes. No, you can't change the number short of modifying the code. > so you would want to break each disk i/o into multiple requests to > maximize its performance (after all the array doesn't have anything > else to do). This is incorrect. You'd still slow things down. Greg -- See complete headers for address, home page and phone numbers finger grog@lemis.com for PGP public key To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19990322092628.V429>