From owner-freebsd-stable Fri Feb 2 17:03:41 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id RAA24730 for stable-outgoing; Fri, 2 Feb 1996 17:03:41 -0800 (PST) Received: from GndRsh.aac.dev.com (GndRsh.aac.dev.com [198.145.92.241]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id RAA24669 for ; Fri, 2 Feb 1996 17:02:01 -0800 (PST) Received: (from rgrimes@localhost) by GndRsh.aac.dev.com (8.6.12/8.6.12) id RAA22138; Fri, 2 Feb 1996 17:00:14 -0800 From: "Rodney W. Grimes" Message-Id: <199602030100.RAA22138@GndRsh.aac.dev.com> Subject: Re: New version of ccd driver available To: bmk@dtr.com Date: Fri, 2 Feb 1996 17:00:13 -0800 (PST) Cc: stable@FreeBSD.ORG In-Reply-To: <199602021914.LAA14322@dtr.com> from "bmk@dtr.com" at Feb 2, 96 11:14:25 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-stable@FreeBSD.ORG Precedence: bulk [Sic to use the list for this, but it may get to you] > > > (7) What's the second field in /etc/ccd.conf? > > > That's the "interleave size". Basically, the ccd driver will write > > this many sectors (usually 512 bytes) to one disk before it moves to > > the next disk. As a special case, a zero here means no interleave, > > i.e., to concatenate disks serially. > > > We have found that in FFS, a value of 16 usually optimizes read > > performance, while the write peaks with a much larger value (like > > 512). This probably has to do with cluster_write() thinking it's > > writing to a single disk when it's actually not. This is one of the > > things we are planning to fix. > > I've done some performance testing - both mirrored and striped - on > one of my systems. I noted a different performance curve than you did, > so I thought you might like to see them. (I used a SCSI-I disk, and > provided the results run on a standard FFS - the numbers are mostly > useful to compare to the FFS baseline results.) > > The results are at "http://www.dtr.com/ccd" - Mostly it's raw results If you are wondering why you have not heard from me I am having problems resolving your domain: Mail Queue (1 request) --Q-ID-- --Size-- -----Q-Time----- ------------Sender/Recipient------------ XAA20111 801 Wed Jan 31 23:21 sales (dtr.com: Name server timeout) bmk@dtr.com GndRsh:rgrimes {166} nslookup Default Server: GndRsh.aac.dev.com Address: 0.0.0.0 > set query=any > dtr.com. Server: GndRsh.aac.dev.com Address: 0.0.0.0 Non-authoritative answer: dtr.com nameserver = NS.TRANSPORT.COM dtr.com nameserver = NS.MLN.COM Authoritative answers can be found from: dtr.com nameserver = NS.TRANSPORT.COM dtr.com nameserver = NS.MLN.COM NS.TRANSPORT.COM internet address = 204.119.17.6 NS.MLN.COM internet address = 199.46.16.1 > server 204.119.17.6 Default Server: www.transport.com Address: 204.119.17.6 > dtr.com. Server: www.transport.com Address: 204.119.17.6 Non-authoritative answer: dtr.com nameserver = mln.mln.com dtr.com preference = 10, mail exchanger = mail.dtr.com Authoritative answers can be found from: dtr.com nameserver = mln.mln.com mail.dtr.com internet address = 204.119.17.5 > a) You have a bad set of deligations for dtr.com b) I can not resolve mln.mln.com c) I can not reach ns.mln.com (199.46.16.1) > and a quick-n-dirty summary and Excel spreadsheet with the performance > curve graphed. The results and summary are in ASCII. I have also > included the script that I used to run the tests. > > I've also got some ambitious ideas for improving the ccd driver - > nothing coded, just some raw ideas. (hot swappable mirrors, etc.). > At work, I deal with three different host-based fault-tolerance > implementations (Sequent Dynix and ptx/SVM, as well as the Solaris disk > suite) - the current ccd driver resembles Dynix; some of the ideas I > have in mind are borrowed from SVM and Solaris. > > If you're interested in hearing what I have in mind, I'll write > something up and forward it to you. > > -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation Company Reliable computers for FreeBSD