From owner-freebsd-hackers Tue Mar 27 0:41:51 2001 Delivered-To: freebsd-hackers@freebsd.org Received: from wantadilla.lemis.com (wantadilla.lemis.com [192.109.197.80]) by hub.freebsd.org (Postfix) with ESMTP id 55A6C37B71B; Tue, 27 Mar 2001 00:41:45 -0800 (PST) (envelope-from grog@lemis.com) Received: by wantadilla.lemis.com (Postfix, from userid 1004) id 113D36A90D; Tue, 27 Mar 2001 18:11:42 +0930 (CST) Date: Tue, 27 Mar 2001 18:11:41 +0930 From: Greg Lehey To: Alfred Perlstein Cc: "Michael C . Wu" , dillon@FreeBSD.ORG, fs@FreeBSD.ORG, hackers@FreeBSD.ORG Subject: Re: Vinum stripe size (was: tuning a VERY heavily (30.0) loaded server) Message-ID: <20010327181141.H1161@wantadilla.lemis.com> References: <20010320111144.A51924@peorth.iteration.net> <20010327171653.G1161@wantadilla.lemis.com> <20010327000503.T9431@fw.wintelcom.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <20010327000503.T9431@fw.wintelcom.net>; from bright@wintelcom.net on Tue, Mar 27, 2001 at 12:05:03AM -0800 Organization: LEMIS, PO Box 460, Echunga SA 5153, Australia Phone: +61-8-8388-8286 Fax: +61-8-8388-8725 Mobile: +61-418-838-708 WWW-Home-Page: http://www.lemis.com/~grog X-PGP-Fingerprint: 6B 7B C3 8C 61 CD 54 AF 13 24 52 F8 6D A4 95 EF Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Tuesday, 27 March 2001 at 0:05:03 -0800, Alfred Perlstein wrote: > * Greg Lehey [010326 23:47] wrote: >> On Tuesday, 20 March 2001 at 11:11:44 -0600, Michael C . Wu wrote: >>> [Lengthy email, bear with me please, it is quite interesting. >>> This box averages 30.0 load with no problems.] >>> >>> >>> >>> Average file size is about 4K. /home/bbsusers* is on a vinum >>> stripe'd volume with 3 Ultra160 9G 10000RPM drives on sym0 at stripe >>> size 256K, Greg: I know this should be a prime number, >> >> No, there's no requirement for it to be a prime number. The only >> problem is that with 32 MB cylinder groups and a power of two stripe >> size and subdisk count, you end up with all the superblocks on one >> subdisk, which is a performance issue. Choose the stripe size so that >> the superblocks are roughly evenly distributed. >> >>> can we safely use <150K stripe sizes? >> >> Safely, yes. But as somebody else has observed, you are probably disk >> I/O bound. Reducing the stripe size will tend to increase the disk >> load, though probably not very much if your files are all 4 kB. I'd >> go for something like 273 kB stripes. > > Do you think it'd be worth it to have vinum carp about what may > be a non optimal stripe size? > > "Warning N is probably a bad idea for a stripe size, see docs" Only if it can recognize the fact correctly. 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-hackers" in the body of the message