Date: Sun, 21 Mar 1999 13:38:09 -0800 (PST) From: Tom <tom@sdf.com> To: Greg Lehey <grog@lemis.com> Cc: Nick Hilliard <nick@iol.ie>, freebsd-scsi@FreeBSD.ORG Subject: Re: dpt raid-5 performance Message-ID: <Pine.BSF.4.05.9903211333100.24283-100000@misery.sdf.com> In-Reply-To: <19990322090015.S429@lemis.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 22 Mar 1999, Greg Lehey wrote: > On Sunday, 21 March 1999 at 12:54:23 -0800, Tom wrote: > > > > On Sun, 21 Mar 1999, Greg Lehey wrote: > > > >>> FreeBSD will likely never send IOs big enough for the strip size of > >>> 512kb to ever be useful. > >> > >> It does. The largest I/O transfer is 64 kB. You don't want to > >> fragment requests, because that increases the I/O load on the array. > > > > At the cost of single-process performance. > > I don't know what you mean here. Can you explain that statement? Small strip sizes are good for single-user situations (like running Bonnie), because the IO load will be split over all drives. Large strip sizes are good for multi-user situations, where the extra overhead of the transactions becomes a problem. DPT starts with a default of 16KB for the strip size. Until recently, it couldn't be increased over 64KB. I've used these cards quite extensively under FreeBSD. Remember what is good for DPT cards is not always what is good for other systems, particularly software solutions. > Greg > -- > See complete headers for address, home page and phone numbers > finger grog@lemis.com for PGP public key Tom 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?Pine.BSF.4.05.9903211333100.24283-100000>