From owner-freebsd-geom@FreeBSD.ORG Fri Jan 30 18:17:51 2009 Return-Path: Delivered-To: freebsd-geom@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 55D71106564A; Fri, 30 Jan 2009 18:17:51 +0000 (UTC) (envelope-from ulf.lilleengen@gmail.com) Received: from bene2.itea.ntnu.no (bene2.itea.ntnu.no [IPv6:2001:700:300:3::57]) by mx1.freebsd.org (Postfix) with ESMTP id 5AA488FC19; Fri, 30 Jan 2009 18:17:50 +0000 (UTC) (envelope-from ulf.lilleengen@gmail.com) Received: from localhost (localhost [127.0.0.1]) by bene2.itea.ntnu.no (Postfix) with ESMTP id AE91F90004; Fri, 30 Jan 2009 19:17:48 +0100 (CET) Received: from carrot (unknown [IPv6:2001:700:300:3::184]) by bene2.itea.ntnu.no (Postfix) with ESMTP id 5671490001; Fri, 30 Jan 2009 19:17:48 +0100 (CET) Date: Fri, 30 Jan 2009 19:18:18 +0000 From: Ulf Lilleengen To: Eric Anderson Message-ID: <20090130191818.GB1275@carrot.BalPol.Local> References: <6612C205-C346-4493-9DA4-3B5A73E9A4F7@freebsd.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <6612C205-C346-4493-9DA4-3B5A73E9A4F7@freebsd.org> User-Agent: Mutt/1.5.18 (2008-05-17) X-Virus-Scanned: Debian amavisd-new at bene2.itea.ntnu.no Cc: freebsd-geom@freebsd.org Subject: Re: Performance numbers? X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list Reply-To: "..."@carrot.BalPol.Local List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 30 Jan 2009 18:17:51 -0000 On Fri, Jan 30, 2009 at 10:40:46AM -0600, Eric Anderson wrote: > Hi GEOMers! > > Does anyone have any benchmarks or numbers relating to GEOM performance? > > I tried doing some on my own, but I didn't get very satisfactory > results, so I'm curious what others have seen or used. > > My hardware is a Core 2 Quad, with 4GB of ram. > > First, I made an mdconfig'ed malloc backed 'disk' of 1.5GB. Then, I > tried running such tools as rawio, and diskinfo. rawio fails with > input/output errors, and diskinfo wants a larger device to give the > full stats. I ended up using purely dd since that worked. > Interestingly enough, dd'ing to the malloc device results in about > 1000 operations per second, regardless of a blocksize of 512bytes or > 1MB. > > Do you all expect this 1000 operations per second to be limited by the > mdconfig'ed device, GEOM, both, all, none, ?? I'm looking for > pointers here mostly. I really would like to see how many operations > (512byte really) the GEOM infrastructure can do on a particular piece > of hardware, so if there is a simpler way to do this test, give me a > hint. :) > Maybe try dtrace to find bottlenecks etc? -- Ulf Lilleengen