From owner-freebsd-performance@FreeBSD.ORG Wed Dec 21 14:19:41 2011 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 100841065673 for ; Wed, 21 Dec 2011 14:19:41 +0000 (UTC) (envelope-from schulra@earlham.edu) Received: from chkenon.earlham.edu (chkenon.earlham.edu [159.28.1.87]) by mx1.freebsd.org (Postfix) with ESMTP id B09FD8FC1B for ; Wed, 21 Dec 2011 14:19:40 +0000 (UTC) X-ASG-Debug-ID: 1324477177-037b9f16aa1707e0001-3XdwJY Received: from tdream.lly.earlham.edu (tdream.lly.earlham.edu [159.28.7.241]) by chkenon.earlham.edu with ESMTP id ZVlfb3N6eiBaM7Jk; Wed, 21 Dec 2011 09:19:37 -0500 (EST) X-Barracuda-Envelope-From: schulra@earlham.edu X-Barracuda-Apparent-Source-IP: 159.28.7.241 Date: Wed, 21 Dec 2011 09:20:59 -0500 (EST) From: Randy Schultz X-X-Sender: schulra@tdream.lly.earlham.edu To: Tom Evans In-Reply-To: X-ASG-Orig-Subj: Re: Benchmark (Phoronix): FreeBSD 9.0-RC2 vs. Oracle Linux 6.1 Server Message-ID: References: <4EE1EAFE.3070408@m5p.com> <4EE2AE64.9060802@m5p.com> <4EE88343.2050302@m5p.com> <4EE933C6.4020209@zedat.fu-berlin.de> <20111215024249.GA13557@icarus.home.lan> <4EE9A2A0.80607@zedat.fu-berlin.de> <4EF1121F.9010209@zedat.fu-berlin.de> <4EF11B57.7090007@phoronix.com> User-Agent: Alpine 2.00 (BSF 1167 2008-08-23) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Barracuda-Connect: tdream.lly.earlham.edu[159.28.7.241] X-Barracuda-Start-Time: 1324477177 X-Barracuda-URL: http://159.28.1.87:8000/cgi-mod/mark.cgi X-Virus-Scanned: by bsmtpd at earlham.edu X-Barracuda-Spam-Score: 0.89 X-Barracuda-Spam-Status: No, SCORE=0.89 using global scores of TAG_LEVEL=1000.0 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=1000.0 tests=SARE_ADLTSUB4 X-Barracuda-Spam-Report: Code version 3.2, rules version 3.2.2.83708 Rule breakdown below pts rule name description ---- ---------------------- -------------------------------------------------- 0.89 SARE_ADLTSUB4 Apparent spam seems to contain porn subject Cc: freebsd-performance@freebsd.org Subject: Re: Benchmark (Phoronix): FreeBSD 9.0-RC2 vs. Oracle Linux 6.1 Server X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 21 Dec 2011 14:19:41 -0000 On Wed, 21 Dec 2011, Tom Evans spaketh thusly: -} -}I think that a good SA will at least consider how drives are arranged. -}We don't just slap ZFS on a single disk and expect magic to happen, we -}consider how write heavy a system will be and consider a dedicated -}ZIL, we consider what proportion of files will be re-read and how much -}application memory will be required and adjust ARC and L2ARC -}accordingly. Tuning and foresight are important. I agree whole-heartedly. I guess I wasn't clear. I wasn't trying to say most SA's never tune, only that from watching other SA's over the years, little tuning is done. IOW, while certainly some tune and play, I suspect that few build up a system then try it with and without hyperthreading, test how big the kern.ipc.shmmax needs to be or if tuning the inflight bits has any effect. Heh, I'ld do this all day long if I could. ;> I love tweaking and tuning, digging into docs and seeing if I can eke a little bit more out of the box. Hmmm, I wonder if the reason base benchmarks like Phoronix have become important is because not a lot of SA's spend a lot of time tuning... Or perhaps they just use it as a starting point. However my main point is that saying something needs to be tuned to get more, or a "proper", speed seems like a slippery slope. "There's always 1 more thing to try." (TM) ;> This then bring up the importance of something previously mentioned - what ever you do, try it yourself in your environment. Do whatever amount of tuning you do (or don't do) and try it. In our environment, fbsd stomps linux for a mail relay. OTOH linux's iSCSI initiator stomps fbsd's. Heh, what would be really cool even if only from an academic perspective would be to take a stock install of OS's, benchmark an app, then tune the them to the max and re-run the tests to see the difference. /me eyes the student SA's... ;> -- Randy (schulra@earlham.edu) 765.983.1283 <*> nosce te ipsum