From owner-freebsd-stable Mon May 1 17:12:41 2000 Delivered-To: freebsd-stable@freebsd.org Received: from mass.cdrom.com (adsl-63-202-176-114.dsl.snfc21.pacbell.net [63.202.176.114]) by hub.freebsd.org (Postfix) with ESMTP id E623D37B6A6 for ; Mon, 1 May 2000 17:12:38 -0700 (PDT) (envelope-from msmith@mass.cdrom.com) Received: from mass.cdrom.com (localhost [127.0.0.1]) by mass.cdrom.com (8.9.3/8.9.3) with ESMTP id RAA04531; Mon, 1 May 2000 17:20:57 -0700 (PDT) (envelope-from msmith@mass.cdrom.com) Message-Id: <200005020020.RAA04531@mass.cdrom.com> X-Mailer: exmh version 2.1.1 10/15/1999 To: dg@root.com Cc: freebsd-stable@FreeBSD.ORG Subject: Re: How good is AMI MegaRAID support? In-reply-to: Your message of "Mon, 01 May 2000 16:43:10 PDT." <200005012343.QAA28681@implode.root.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 01 May 2000 17:20:57 -0700 From: Mike Smith Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > >> How fast is fast enough? Do you need a disk subsystem that can > >> handle bursts of ~2500 USENET news history database lookups per > >> second? What would it take to get to that level? Do you need to see > >> rawio read performance (average tranfer size of one sector w/ 64 > >> simultaneous processes) of around 5,000 read operations per second? > > > >We're running quite a bit more than 5,000 iops/sec on the EMC Symmetrix > >box that's replacing the A-3500s. > > > >Just depends on what it is you need done. > > TeraSolutions' RAID systems (TSR-2200) as used on ftp.freesoftware.com are > capable of >9,000 IOPS. Is this a controller capability, or does it represent a sustainable load over the entire array? How do you measure this? (I'd love to add to my benchmark/test suite). FWIW, most of the low-end PCI:SCSI RAID controllers claim throughput in the 3-5k IOPs, and 20k is not an uncommon claim for mid-high end controllers. Simon Shapiro was pushing over 20k on the DPT Century adapters in "real" applications. I've had a hard time generating more than 3k or so out of a FreeBSD box's I/O subsystem - we cluster so aggressively that I typically run out of I/O bandwidth before I hit an IOP limit. > There are other issues to consider when thinking about software RAID-5. The > most important of these is the (lack of) non-volatile write-back cache. This is definitely a kicker. The better controllers will let you swap the BBU and RAM into a new controller too, eg. if you've fried a SCSI channel, and still preserve your unwritten data. (Mylex and Adaptec, at least, do this.) -- \\ Give a man a fish, and you feed him for a day. \\ Mike Smith \\ Tell him he should learn how to fish himself, \\ msmith@freebsd.org \\ and he'll hate you for a lifetime. \\ msmith@cdrom.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message