From owner-freebsd-performance@FreeBSD.ORG Sun Dec 5 20:44:35 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 63FDC16A504 for ; Sun, 5 Dec 2004 20:44:35 +0000 (GMT) Received: from web14105.mail.yahoo.com (web14105.mail.yahoo.com [216.136.172.135]) by mx1.FreeBSD.org (Postfix) with SMTP id 1D76143D68 for ; Sun, 5 Dec 2004 20:44:35 +0000 (GMT) (envelope-from cguttesen@yahoo.dk) Received: (qmail 9343 invoked by uid 60001); 5 Dec 2004 20:44:34 -0000 Message-ID: <20041205204434.9341.qmail@web14105.mail.yahoo.com> Received: from [194.248.174.50] by web14105.mail.yahoo.com via HTTP; Sun, 05 Dec 2004 21:44:34 CET Date: Sun, 5 Dec 2004 21:44:34 +0100 (CET) From: Claus Guttesen To: Sean Chittenden In-Reply-To: <6174636A-43BC-11D9-ACAF-000A95C705DC@chittenden.org> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit cc: freebsd-performance@freebsd.org Subject: Re: postgresql on FreeBSD 5.3 and high load X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 05 Dec 2004 20:44:35 -0000 > Have you looked to see if you're using spin locks or > not? Search for > "amd64 x86_64 spin lock" in PostgreSQL's archives. > I think spin locks > were disabled for amd64. When you're at a load of > 70, look in > pg_catalog.pg_locks to see if you have any ungranted > locks. Thank you for your input. Searched postgresl.org's archives, both wihtin and via google, but only found information that related primarily to Linux. > > Also, is your application primarily read? You may > want to investigate > using pgpool as a way of reducing the overhead for > connection startup. There are more reads than writes, but I haven't found the exact numbers (yet). > Another option is to use memcached and remove some > queries from the > database all together. -sc I will take a look at these utilities, but will wait until January next year, where the load have decreased. Right now I'll throw some more hardware at it, a quad-opteron which was planned (in May next year), and will go through the php-code to optimize. I increased the shared_buffers to 65536, but it did not change much. I'll decrease it to 16384. Read a thread related to 'mbwait postgresql freebsd' which I found at freebsd's site. This suggested that a (too) high value could be counterproductive. regards Claus From owner-freebsd-performance@FreeBSD.ORG Sun Dec 5 20:58:51 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 9379816A52B for ; Sun, 5 Dec 2004 20:58:51 +0000 (GMT) Received: from mail.trippynames.com (mail.trippynames.com [38.113.223.19]) by mx1.FreeBSD.org (Postfix) with ESMTP id 6D8BC43D1D for ; Sun, 5 Dec 2004 20:58:51 +0000 (GMT) (envelope-from sean@chittenden.org) Received: from localhost (localhost [127.0.0.1]) by mail.trippynames.com (Postfix) with ESMTP id 8D363A6C21; Sun, 5 Dec 2004 12:58:50 -0800 (PST) Received: from mail.trippynames.com ([127.0.0.1]) by localhost (rand.nxad.com [127.0.0.1]) (amavisd-new, port 10024) with LMTP id 31841-05; Sun, 5 Dec 2004 12:58:49 -0800 (PST) Received: from [192.168.1.4] (dsl081-069-073.sfo1.dsl.speakeasy.net [64.81.69.73]) by mail.trippynames.com (Postfix) with ESMTP id 1E485A6C3A; Sun, 5 Dec 2004 12:58:48 -0800 (PST) In-Reply-To: <20041205204434.9341.qmail@web14105.mail.yahoo.com> References: <20041205204434.9341.qmail@web14105.mail.yahoo.com> Mime-Version: 1.0 (Apple Message framework v619) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Message-Id: <72207B58-4700-11D9-9EB9-000A95C705DC@chittenden.org> Content-Transfer-Encoding: 7bit From: Sean Chittenden Date: Sun, 5 Dec 2004 12:58:42 -0800 To: Claus Guttesen X-Mailer: Apple Mail (2.619) cc: freebsd-performance@freebsd.org Subject: Re: postgresql on FreeBSD 5.3 and high load X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 05 Dec 2004 20:58:51 -0000 >> Have you looked to see if you're using spin locks or >> not? Search for "amd64 x86_64 spin lock" in PostgreSQL's archives. >> I think spin locks were disabled for amd64. When you're at a load of >> 70, look in pg_catalog.pg_locks to see if you have any ungranted >> locks. > > Thank you for your input. > > Searched postgresl.org's archives, both wihtin and via > google, but only found information that related > primarily to Linux. Before I fired that off, I should've asked, if you do: SELECT * FROM pg_catalog.pg_locks WHERE granted = FALSE Are you getting any rows? If not, slow locks aren't the issue. Have you run iostat(1) to see if you're saturating your IO? >> Also, is your application primarily read? You may >> want to investigate >> using pgpool as a way of reducing the overhead for >> connection startup. > > There are more reads than writes, but I haven't found > the exact numbers (yet). Look in pg_catalog.pg_stat* for an idea of your workload. http://developer.postgresql.org/docs/postgres/monitoring- stats.html#MONITORING-STATS-VIEWS-TABLE If you can come back with a more specific problem other than "load is high," it's much easier to get more precise help regarding whatever the problem is. A load of 70 may just mean you've got a damn busy database. :) -sc -- Sean Chittenden From owner-freebsd-performance@FreeBSD.ORG Mon Dec 6 19:31:48 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id B460416A4D0 for ; Mon, 6 Dec 2004 19:31:48 +0000 (GMT) Received: from web14121.mail.yahoo.com (web14121.mail.yahoo.com [66.163.171.112]) by mx1.FreeBSD.org (Postfix) with SMTP id 9542743D1D for ; Mon, 6 Dec 2004 19:31:48 +0000 (GMT) (envelope-from cguttesen@yahoo.dk) Received: (qmail 85707 invoked by uid 60001); 6 Dec 2004 19:31:48 -0000 Message-ID: <20041206193148.85705.qmail@web14121.mail.yahoo.com> Received: from [194.248.174.58] by web14121.mail.yahoo.com via HTTP; Mon, 06 Dec 2004 20:31:48 CET Date: Mon, 6 Dec 2004 20:31:48 +0100 (CET) From: Claus Guttesen To: Sean Chittenden In-Reply-To: <72207B58-4700-11D9-9EB9-000A95C705DC@chittenden.org> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit cc: freebsd-performance@freebsd.org Subject: Re: postgresql on FreeBSD 5.3 and high load (solved) X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 06 Dec 2004 19:31:48 -0000 Problem solved. It's a bit embarassing, but it turned out that it was a script which caused the high load. The script did unfortunately not bail out when a required condition was not satisfied. It queried the postgresql-server, and one thing led to the other. I (re)wrote the script, and it's now working as intended. Sorry for the noise. Claus From owner-freebsd-performance@FreeBSD.ORG Tue Dec 7 17:15:42 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id D7A7016A4CE for ; Tue, 7 Dec 2004 17:15:42 +0000 (GMT) Received: from sb.santaba.com (sb.santaba.com [207.154.84.26]) by mx1.FreeBSD.org (Postfix) with ESMTP id ADA1E43D78 for ; Tue, 7 Dec 2004 17:15:42 +0000 (GMT) (envelope-from jbehl@fastclick.com) Received: from [192.168.3.100] (unknown [205.180.85.193]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by sb.santaba.com (Postfix) with ESMTP id 071D828433; Tue, 7 Dec 2004 09:15:42 -0800 (PST) Message-ID: <41B5E5A4.6060407@fastclick.com> Date: Tue, 07 Dec 2004 09:17:24 -0800 From: Jeff Behl User-Agent: Mozilla Thunderbird 0.8 (Windows/20040913) X-Accept-Language: en-us, en MIME-Version: 1.0 To: Sean Chittenden References: <41AE651F.3040104@fastclick.com> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit cc: freebsd-performance@freebsd.org Subject: Re: %cpu in system - squid performance in FreeBSD 5.3 X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 07 Dec 2004 17:15:43 -0000 I upgraded to STABLE but most cpu time is still being spent in system. This system is doing ~20Mb/s total with all content being grabbed out of memory. I see similar results when running MySQL (a lot of time being spent in system) Any ideas on what updates to be on the lookout for that might help with this? Am I right in guessing that this is a SMP issue and doesn't have anything to do with AMD architecture? thx FreeBSD www2 5.3-STABLE FreeBSD 5.3-STABLE #2: Sun Dec 5 21:06:14 PST 2004 root@www2.cdn.sjc:/usr/obj/usr/src/sys/SMP amd64 last pid: 15702; load averages: 0.15, 0.31, 0.31 up 0+19:55:14 09:09:28 38 processes: 2 running, 36 sleeping CPU states: 5.4% user, 0.0% nice, 12.7% system, 3.4% interrupt, 78.4% idle Mem: 163M Active, 284M Inact, 193M Wired, 72K Cache, 214M Buf, 1245M Free Swap: 4069M Total, 4069M Free PID USERNAME PRI NICE SIZE RES STATE C TIME WCPU CPU COMMAND 486 squid 96 0 79820K 73996K CPU1 1 110:00 15.04% 15.04% squid 480 squid 96 0 75804K 70012K select 0 105:56 14.89% 14.89% squid Sean Chittenden wrote: >> but the % system time can fluctuate up to 60 at times. My question is >> if this is about the type of performance I could expect, or if people >> have seen better. > > > I don't know about other people, but I suspect you're running into > lock contention. Try using a post 5.3 snapshot (something from > RELENG_5) since alc@ has set debug.mpsafevm=1, which lets many calls > to the VM run without GIANT, which I suspect is your problem and why > the system usage is all over the place. -sc > From owner-freebsd-performance@FreeBSD.ORG Wed Dec 8 12:36:47 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 2FFC816A4CE for ; Wed, 8 Dec 2004 12:36:47 +0000 (GMT) Received: from server1.carmatec.com (server1.carmatec.com [66.45.229.226]) by mx1.FreeBSD.org (Postfix) with ESMTP id 8F5DD43D5A for ; Wed, 8 Dec 2004 12:36:46 +0000 (GMT) (envelope-from akhthar@carmatec.com) Received: from [61.95.203.89] (helo=192.168.0.58) by server1.carmatec.com with esmtpa (Exim 4.43) id 1Cc13k-00013r-1l for freebsd-performance@freebsd.org; Wed, 08 Dec 2004 07:36:22 -0500 From: "Akhthar Parvez. K" Organization: Carmatec Solutions To: freebsd-performance@freebsd.org Date: Wed, 8 Dec 2004 18:08:13 +0530 User-Agent: KMail/1.5 MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200412081808.13353.akhthar@carmatec.com> X-AntiAbuse: This header was added to track abuse, please include it with any abuse report X-AntiAbuse: Primary Hostname - server1.carmatec.com X-AntiAbuse: Original Domain - freebsd.org X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12] X-AntiAbuse: Sender Address Domain - carmatec.com X-Source: X-Source-Args: X-Source-Dir: Subject: Memory issues:-( X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list Reply-To: akhthar@carmatec.com List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 08 Dec 2004 12:36:47 -0000 Hi All, I am facing a lot of memory issues with all services in the server. Following log will tell you what is the exact problem. dmesg 18575 (exim-4.43-0), uid 26: exited on signal 11 pid 18595 (exim-4.43-0), uid 26: exited on signal 11 pid 18512 (httpd), uid 65534: exited on signal 11 pid 18531 (httpd), uid 65534: exited on signal 10 pid 18518 (httpd), uid 65534: exited on signal 10 pid 18516 (httpd), uid 65534: exited on signal 11 pid 18524 (httpd), uid 65534: exited on signal 11 pid 18511 (httpd), uid 65534: exited on signal 4 pid 18496 (httpd), uid 65534: exited on signal 11 pid 18494 (httpd), uid 65534: exited on signal 11 pid 18519 (httpd), uid 65534: exited on signal 10 pid 18532 (httpd), uid 65534: exited on signal 11 pid 18521 (httpd), uid 65534: exited on signal 10 pid 18510 (httpd), uid 65534: exited on signal 10 pid 18614 (httpd), uid 65534: exited on signal 11 pid 18527 (httpd), uid 65534: exited on signal 10 pid 18722 (exim-4.43-0), uid 26: exited on signal 11 pid 18806 (exim-4.43-0), uid 26: exited on signal 11 pid 18728 (exim-4.43-0), uid 26: exited on signal 11 pid 18964 (exim-4.43-0), uid 26: exited on signal 11 pid 18998 (exim-4.43-0), uid 26: exited on signal 11 pid 19004 (exim-4.43-0), uid 26: exited on signal 11 pid 19008 (exim-4.43-0), uid 26: exited on signal 11 pid 19061 (exim-4.43-0), uid 26: exited on signal 11 pid 19067 (exim-4.43-0), uid 26: exited on signal 11 pid 19073 (exim-4.43-0), uid 26: exited on signal 11 pid 18874 (exim-4.43-0), uid 26: exited on signal 11 pid 19100 (exim-4.43-0), uid 26: exited on signal 11 pid 19106 (exim-4.43-0), uid 26: exited on signal 11 pid 19115 (exim-4.43-0), uid 26: exited on signal 11 pid 19122 (exim-4.43-0), uid 26: exited on signal 11 pid 19149 (exim-4.43-0), uid 26: exited on signal 11 pid 18617 (httpd), uid 65534: exited on signal 11 pid 19161 (exim-4.43-0), uid 26: exited on signal 11 pid 19165 (exim-4.43-0), uid 26: exited on signal 11 pid 19175 (exim-4.43-0), uid 26: exited on signal 11 pid 18505 (httpd), uid 65534: exited on signal 11 pid 18691 (httpd), uid 65534: exited on signal 11 pid 19224 (exim-4.43-0), uid 26: exited on signal 11 pid 19145 (exim-4.43-0), uid 26: exited on signal 11 pid 19275 (exim-4.43-0), uid 26: exited on signal 11 pid 19302 (exim-4.43-0), uid 26: exited on signal 11 pid 19307 (exim-4.43-0), uid 26: exited on signal 11 pid 19309 (exim-4.43-0), uid 26: exited on signal 11 pid 19316 (exim-4.43-0), uid 26: exited on signal 11 pid 19220 (exim-4.43-0), uid 26: exited on signal 11 pid 19348 (exim-4.43-0), uid 26: exited on signal 11 pid 19376 (exim-4.43-0), uid 26: exited on signal 11 pid 19380 (exim-4.43-0), uid 26: exited on signal 11 pid 19388 (exim-4.43-0), uid 26: exited on signal 11 pid 19435 (exim-4.43-0), uid 26: exited on signal 11 pid 19323 (exim-4.43-0), uid 26: exited on signal 11 pid 19482 (exim-4.43-0), uid 26: exited on signal 11 pid 19496 (exim-4.43-0), uid 26: exited on signal 11 pid 19510 (exim-4.43-0), uid 26: exited on signal 11 pid 19600 (exim-4.43-0), uid 26: exited on signal 11 Anyone has any idea what might caused the issue. Following is the details abt memory. vmstat procs memory page disks faults cpu r b w avm fre flt re pi po fr sr da0 md0 in sy cs us sy id 9 0 0 2653476 201136 3014 4 2 0 2926 37 0 0 785 67046 5994 25 30 45 top 141 processes: 4 running, 137 sleeping CPU states: 14.2% user, 0.0% nice, 85.6% system, 0.2% interrupt, 0.0% idle Mem: 842M Active, 2078M Inact, 375M Wired, 156M Cache, 199M Buf, 36M Free Swap: 2048M Total, 32K Used, 2048M Free Any help will be greatly appreciated. -- With Regards, Akhthar Parvez.K From owner-freebsd-performance@FreeBSD.ORG Wed Dec 8 16:26:40 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id E6E8A16A4CE for ; Wed, 8 Dec 2004 16:26:40 +0000 (GMT) Received: from CPE000103d44c07-CM000f9f7ae88c.cpe.net.cable.rogers.com (CPE000103d44c07-CM000f9f7ae88c.cpe.net.cable.rogers.com [69.193.222.195]) by mx1.FreeBSD.org (Postfix) with ESMTP id B9F9243D41 for ; Wed, 8 Dec 2004 16:26:40 +0000 (GMT) (envelope-from mikej@rogers.com) Received: from cpe000103d44c07-cm000f9f7ae88c.cpe.net.cable.rogers.com (localhost [127.0.0.1])with ESMTP id 0D8C62957CF; Wed, 8 Dec 2004 11:26:30 -0500 (EST) Received: from 207.219.213.163 (proxying for unknown) (SquirrelMail authenticated user mikej); by cpe000103d44c07-cm000f9f7ae88c.cpe.net.cable.rogers.com with HTTP; Wed, 8 Dec 2004 11:26:30 -0500 (EST) Message-ID: <34687.207.219.213.163.1102523190.squirrel@207.219.213.163> In-Reply-To: <200412081808.13353.akhthar@carmatec.com> References: <200412081808.13353.akhthar@carmatec.com> Date: Wed, 8 Dec 2004 11:26:30 -0500 (EST) From: "Mike Jakubik" To: akhthar@carmatec.com User-Agent: SquirrelMail/1.4.3a X-Mailer: SquirrelMail/1.4.3a MIME-Version: 1.0 Content-Type: text/plain;charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Priority: 3 (Normal) Importance: Normal X-wettoast-MailScanner-Information: Please contact the ISP for more information X-wettoast-MailScanner: Found to be clean X-MailScanner-From: mikej@rogers.com cc: freebsd-performance@freebsd.org Subject: Re: Memory issues:-( X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 08 Dec 2004 16:26:41 -0000 Akhthar Parvez. K said: > Hi All, > > I am facing a lot of memory issues with all services in the server. > Following log will tell you what is the exact problem. So.. replace the memory. Sig 11 is usually a sign of bad hardware. I don't see how this belongs in the performance mail list, perhaps email hardware or questions in the future. From owner-freebsd-performance@FreeBSD.ORG Wed Dec 8 16:51:23 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 1B6AA16A4CE for ; Wed, 8 Dec 2004 16:51:23 +0000 (GMT) Received: from tardis.uts.ohio-state.edu (tardis.uts.ohio-state.edu [128.146.116.116]) by mx1.FreeBSD.org (Postfix) with ESMTP id C6B1B43D5D for ; Wed, 8 Dec 2004 16:51:22 +0000 (GMT) (envelope-from anderson@wks.uts.ohio-state.edu) Received: from lennier.uts.ohio-state.edu (lennier.uts.ohio-state.edu [128.146.116.7])iB8GpMhi008235 for ; Wed, 8 Dec 2004 11:51:22 -0500 (EST) Date: Wed, 8 Dec 2004 11:51:21 -0500 (EST) From: Bill Anderson X-X-Sender: To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: 2120S poor performance X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 08 Dec 2004 16:51:23 -0000 I've got a 3x72g RAID5 array with U320 disks on an Adaptec 2120S controller under Freebsd 4.10. I'm getting about 25MB/s for sequential reads/writes (e.g. dd if=/dev/aacd0s1b of=/dev/null bs=32k). I've turned on write caching for the container (container set cache /write_cache_enable) and turned off the read caching, per previous posts to this list. I'm going to try setting hw.aac.iosize_max to 96k tonight to see if that helps. I'm currently using an IDE disk that gets about 45MB/s, so I can hardly consider it an "upgrade" to switch to a 25MB/s U320 SCSI system. :/ I've also read that the 2120S is slow because of its design. I'm trying to figure out whether the performance can be increased significantly (I saw a posting of a linux user getting 37.5MB/s, which although still slow might be acceptable), or if I'm better off getting a new card. I'm thinking the latter is the case based on previous postings, but since it will probably be quite expensive to replace it, I wanted to get some more data. What performance should I be expecting from a decent U320 RAID5 controller? Has anyone gotten a 2120S to perform above 30Mb/s in FreeBSD? What's the cheapest controller that still gives reasonable performance? (If you could give a couple different ones, with their associated performance (under FreeBSD), or tell me where to find such information, that would be great) Thanks, Bill From owner-freebsd-performance@FreeBSD.ORG Thu Dec 9 17:06:56 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id D985616A4CE for ; Thu, 9 Dec 2004 17:06:56 +0000 (GMT) Received: from mail.geeks.org (jacobs.Geeks.ORG [204.153.247.1]) by mx1.FreeBSD.org (Postfix) with ESMTP id 2BA7143D39 for ; Thu, 9 Dec 2004 17:06:56 +0000 (GMT) (envelope-from drechsau@Geeks.ORG) Received: by mail.geeks.org (Postfix, from userid 400) id 781DB20B55; Thu, 9 Dec 2004 11:06:55 -0600 (CST) Date: Thu, 9 Dec 2004 11:06:55 -0600 From: Mike Horwath To: Bill Anderson Message-ID: <20041209170655.GH45512@octanews.net> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: X-PGP-Fingerprint: D8 24 CC E6 47 5F E4 60 BF B7 6E FA BF C7 6E C5 X-GPG-Fingerprint: 6A89 E78A B8B1 69D9 8CDB E966 4A5A C3F9 A1B0 C381 User-Agent: Mutt/1.5.6i cc: freebsd-performance@freebsd.org Subject: Re: 2120S poor performance X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 09 Dec 2004 17:06:57 -0000 On Wed, Dec 08, 2004 at 11:51:21AM -0500, Bill Anderson wrote: > I've got a 3x72g RAID5 array with U320 disks on an Adaptec 2120S > controller under Freebsd 4.10. I'm getting about 25MB/s for sequential > reads/writes (e.g. dd if=/dev/aacd0s1b of=/dev/null bs=32k). I've > turned on write caching for the container (container set cache > /write_cache_enable) and turned off the read caching, per previous posts > to this list. I'm going to try setting hw.aac.iosize_max to 96k tonight > to see if that helps. I'm currently using an IDE disk that gets about > 45MB/s, so I can hardly consider it an "upgrade" to switch to a 25MB/s > U320 SCSI system. :/ > > I've also read that the 2120S is slow because of its design. I'm trying > to figure out whether the performance can be increased significantly (I > saw a posting of a linux user getting 37.5MB/s, which although still slow > might be acceptable), or if I'm better off getting a new card. I'm > thinking the latter is the case based on previous postings, but since it > will probably be quite expensive to replace it, I wanted to get some more > data. > > What performance should I be expecting from a decent U320 RAID5 > controller? > > Has anyone gotten a 2120S to perform above 30Mb/s in FreeBSD? > > What's the cheapest controller that still gives reasonable performance? > (If you could give a couple different ones, with their associated > performance (under FreeBSD), or tell me where to find such information, > that would be great) I know I left quite a bit of cruft above, but I don't see these kinds of performance numbers you are seeing. First, SCSI does a lot with overlapping I/O and tagged queuing, only the newest IDE (SATA) systems are starting to support such things. Second, SCSI disks run at a faster RPM most of the time (only the WD Raptor hits the 10K mark), lowering seek time significantly. Third, my numbers: 31MB/sec using dd if=/dev/aacd0s1b of=/dev/null bs=32k count=20000 34MB/sec using dd if=/dev/aacd0s1b of=/dev/null bs=64k count=20000 33MB/sec using dd if=/dev/aacd0s1b of=/dev/null bs=128k count=20000 Now, go and do some testing with random reads and writes and you will see a big difference between your standard IDE disk and this RAID system. The system I was tesing with was loaded down with MySQL operating with 320 tables open, 110 connections, and 128 queries per second, including one table with 9.7M records and 3.2GB in size. The system only has 2GB of RAM. I use the aac driver for a few of my news feeder boxes, I do primary storage of my history on a mounted md but I also do all of my other work via an aac driven RAID5 partition for the things I worry about. The spool is on an external system and doesn't count :) (the external system reaches upwards of 70MB/sec using the test above via 1Gbps FC, and is storaged by WD Raptor SATA 74GB disks). Each of these boxes handles 1.4TB of inbound news per day, and each have been doing 2TB outbound...the RAID systems have been fine... I also use the amrd driver on my Dell frontend systems for news using RAID5 and 3x73GB 10K disks, with upwards of 320-450 connections per server, no issues from the disk subsystems... Again, random reads and writes are what you might want to worry about, sequential R/W doesn't happen in a real work environment, at least not for me. -- Mike Horwath, reachable via drechsau@Geeks.ORG From owner-freebsd-performance@FreeBSD.ORG Thu Dec 9 18:50:58 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id C04A716A4CE for ; Thu, 9 Dec 2004 18:50:58 +0000 (GMT) Received: from smtp006.bizmail.sc5.yahoo.com (smtp006.bizmail.sc5.yahoo.com [66.163.175.83]) by mx1.FreeBSD.org (Postfix) with SMTP id 9F2AD43D4C for ; Thu, 9 Dec 2004 18:50:58 +0000 (GMT) (envelope-from noackjr@alumni.rice.edu) Received: from unknown (HELO optimator.noacks.org) (noackjr@supercrime.org@70.240.198.174 with login) by smtp006.bizmail.sc5.yahoo.com with SMTP; 9 Dec 2004 18:50:58 -0000 Received: from localhost (localhost [127.0.0.1]) by optimator.noacks.org (Postfix) with ESMTP id A864261B9; Thu, 9 Dec 2004 12:50:57 -0600 (CST) Received: from optimator.noacks.org ([127.0.0.1]) by localhost (optimator.noacks.org [127.0.0.1]) (amavisd-new, port 10024) with LMTP id 03280-02; Thu, 9 Dec 2004 12:50:55 -0600 (CST) Received: from www.noacks.org (localhost [127.0.0.1]) by optimator.noacks.org (Postfix) with ESMTP id 8B9AF614E; Thu, 9 Dec 2004 12:50:55 -0600 (CST) Received: from 69.53.57.66 (SquirrelMail authenticated user noackjr); by www.noacks.org with HTTP; Thu, 9 Dec 2004 12:50:55 -0600 (CST) Message-ID: <17418.69.53.57.66.1102618255.squirrel@69.53.57.66> In-Reply-To: References: Date: Thu, 9 Dec 2004 12:50:55 -0600 (CST) From: "Jon Noack" To: "Bill Anderson" User-Agent: SquirrelMail/1.4.3a X-Mailer: SquirrelMail/1.4.3a MIME-Version: 1.0 Content-Type: text/plain;charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Priority: 3 (Normal) Importance: Normal X-Virus-Scanned: amavisd-new at noacks.org cc: freebsd-performance@freebsd.org Subject: Re: 2120S poor performance X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list Reply-To: noackjr@alumni.rice.edu List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 09 Dec 2004 18:50:58 -0000 Bill Anderson wrote: > I've got a 3x72g RAID5 array with U320 disks on an Adaptec 2120S > controller under Freebsd 4.10. I'm getting about 25MB/s for sequential > reads/writes (e.g. dd if=/dev/aacd0s1b of=/dev/null bs=32k). I've > turned on write caching for the container (container set cache > /write_cache_enable) and turned off the read caching, per previous posts > to this list. I'm going to try setting hw.aac.iosize_max to 96k tonight > to see if that helps. I'm currently using an IDE disk that gets about > 45MB/s, so I can hardly consider it an "upgrade" to switch to a 25MB/s > U320 SCSI system. :/ > > I've also read that the 2120S is slow because of its design. I'm trying > to figure out whether the performance can be increased significantly (I > saw a posting of a linux user getting 37.5MB/s, which although still slow > might be acceptable), or if I'm better off getting a new card. I'm > thinking the latter is the case based on previous postings, but since it > will probably be quite expensive to replace it, I wanted to get some more > data. > > What performance should I be expecting from a decent U320 RAID5 > controller? > > Has anyone gotten a 2120S to perform above 30Mb/s in FreeBSD? > > What's the cheapest controller that still gives reasonable performance? > (If you could give a couple different ones, with their associated > performance (under FreeBSD), or tell me where to find such information, > that would be great) The asr driver is in need of a lot of attention. Still, testing random read/write is much better than testing sequential i/o as most real-world workloads aren't sequential in nature. The raidtest utility (now in ports) does an excellent job with random read/write testing. There are many quality, (relatively) cheap controllers out there, but if you're looking to stick with Adaptec get a 2130S. The 2130S uses the aac driver, which is quite well tuned and maintained by Scott Long. Jon From owner-freebsd-performance@FreeBSD.ORG Thu Dec 9 19:03:30 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 51EED16A4CE for ; Thu, 9 Dec 2004 19:03:30 +0000 (GMT) Received: from multiplay.co.uk (www1.multiplay.co.uk [212.42.16.7]) by mx1.FreeBSD.org (Postfix) with ESMTP id 922FC43D2D for ; Thu, 9 Dec 2004 19:03:29 +0000 (GMT) (envelope-from killing@multiplay.co.uk) Received: from stevenp4 ([193.123.241.40]) by multiplay.co.uk (multiplay.co.uk [212.42.16.7]) (MDaemon.PRO.v7.2.1.R) with ESMTP id md50000783293.msg for ; Thu, 09 Dec 2004 18:54:56 +0000 Message-ID: <023a01c4de21$bae392e0$7f06000a@int.mediasurface.com> From: "Steven Hartland" To: , "Bill Anderson" References: <17418.69.53.57.66.1102618255.squirrel@69.53.57.66> Date: Thu, 9 Dec 2004 19:02:53 -0000 MIME-Version: 1.0 Content-Type: text/plain; format=flowed; charset="iso-8859-1"; reply-type=original Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2900.2180 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.2180 X-Spam-Processed: multiplay.co.uk, Thu, 09 Dec 2004 18:54:56 +0000 (not processed: message from valid local sender) X-MDRemoteIP: 193.123.241.40 X-Return-Path: killing@multiplay.co.uk X-MDaemon-Deliver-To: freebsd-performance@freebsd.org X-MDAV-Processed: multiplay.co.uk, Thu, 09 Dec 2004 18:54:57 +0000 cc: freebsd-performance@freebsd.org Subject: Re: 2120S poor performance X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 09 Dec 2004 19:03:30 -0000 ----- Original Message ----- From: "Jon Noack" > The asr driver is in need of a lot of attention. Still, testing random > read/write is much better than testing sequential i/o as most real-world > workloads aren't sequential in nature. The raidtest utility (now in > ports) does an excellent job with random read/write testing. As a point of note 2015S performs as poor under windows as it does under freebsd so Im not so sure that its totally down to drivers. > There are many quality, (relatively) cheap controllers out there, but if > you're looking to stick with Adaptec get a 2130S. The 2130S uses the aac > driver, which is quite well tuned and maintained by Scott Long. ================================================ This e.mail is private and confidential between Multiplay (UK) Ltd. and the person or entity to whom it is addressed. In the event of misdirection, the recipient is prohibited from using, copying, printing or otherwise disseminating it or any information contained in it. In the event of misdirection, illegible or incomplete transmission please telephone (023) 8024 3137 or return the E.mail to postmaster@multiplay.co.uk. From owner-freebsd-performance@FreeBSD.ORG Thu Dec 9 19:10:03 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 5EC0116A4CF for ; Thu, 9 Dec 2004 19:10:03 +0000 (GMT) Received: from server1.carmatec.com (server1.carmatec.com [66.45.229.226]) by mx1.FreeBSD.org (Postfix) with ESMTP id DFC2543D58 for ; Thu, 9 Dec 2004 19:10:02 +0000 (GMT) (envelope-from akhthar@carmatec.com) Received: from [61.95.203.89] (helo=ws15.carmatec.com) by server1.carmatec.com with esmtpa (Exim 4.43) id 1CcTfp-00080V-OT; Thu, 09 Dec 2004 14:09:36 -0500 Received: from [216.136.204.119] (helo=mx2.freebsd.org) by server1.carmatec.com with esmtp (Exim 4.43) id 1CcTTB-0004Bz-3k for akhthar@carmatec.com; Thu, 09 Dec 2004 13:56:31 -0500 Received: from hub.freebsd.org (hub.freebsd.org [216.136.204.18]) by mx2.freebsd.org (Postfix) with ESMTP id 8B64C56794; Thu, 9 Dec 2004 18:56:43 +0000 (GMT) (envelope-from owner-freebsd-questions@freebsd.org) Received: from hub.freebsd.org (localhost [127.0.0.1]) by hub.freebsd.org (Postfix) with ESMTP id 40D6B16A4DB; Thu, 9 Dec 2004 18:56:40 +0000 (GMT) Delivered-To: freebsd-questions@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id C3A5E16A4CE for ; Thu, 9 Dec 2004 18:56:34 +0000 (GMT) Received: from pi.codefab.com (pi.codefab.com [199.103.21.227]) by mx1.FreeBSD.org (Postfix) with ESMTP id 6D5FD43D64 for ; Thu, 9 Dec 2004 18:56:34 +0000 (GMT) (envelope-from cswiger@mac.com) Received: from [192.168.1.250] (pool-68-161-115-118.ny325.east.verizon.net [68.161.115.118]) by pi.codefab.com (8.12.11/8.12.11) with ESMTP id iB9IuBgb011547 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Thu, 9 Dec 2004 13:56:14 -0500 (EST) Date: Thu, 9 Dec 2004 14:11:46 -0500 From: "Akhthar Parvez. K" Organization: Carmatec Solutions User-Agent: KMail/1.5 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Olivier Nicole References: <200412090049.iB90ntMx029318@mail.cs.ait.ac.th> In-Reply-To: <200412090049.iB90ntMx029318@mail.cs.ait.ac.th> X-Enigmail-Version: 0.86.1.0 X-Enigmail-Supports: pgp-inline, pgp-mime X-Spam-Status: No, hits=-4.5 required=5.5 tests=AWL,BAYES_00 autolearn=ham version=2.64 X-Spam-Checker-Version: SpamAssassin 2.64 (2004-01-11) on pi.codefab.com X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list Sender: owner-freebsd-questions@freebsd.org Errors-To: owner-freebsd-questions@freebsd.org Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200412100041.46272.akhthar@carmatec.com> X-AntiAbuse: This header was added to track abuse, please include it with any abuse report X-AntiAbuse: Primary Hostname - server1.carmatec.com X-AntiAbuse: Original Domain - freebsd.org X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12] X-AntiAbuse: Sender Address Domain - carmatec.com X-Source: X-Source-Args: X-Source-Dir: cc: freebsd-performance@freebsd.org cc: freebsd-questions@freebsd.org Subject: Re: Memory problems:-( X-BeenThere: freebsd-performance@freebsd.org Reply-To: akhthar@carmatec.com List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 09 Dec 2004 19:10:03 -0000 hello, I have even tried replacing the RAM and upgrading the RAM. But nothing worked. And I am facing this issue on both my FreeBSd servers. And no problems in my redhat servers. Please do not give me an unmatured reply as my friends Oliver and Check gave. Olivier Nicole wrote: >> I am facing a lot of memory issues with all services in the server. >> Following log will tell you what is the exact problem. > > Have you tried cleaning the contacts on the memory stick? > > Cotton bud and rubing alchool, thooth paste, sliver cleaning paste, > car polish... OK use alchool last to rinse the contacts... That can > help. Especially is the machine has to deal with some difficult > environment, has been moved recently, etc. This is not a bad thought, although simply removing the RAM and reinserting it properly is often enough to reset the memory and wipe the contacts clean of any minor oxidation on the contacts. Toothpaste or other abrasives are excessive, although the notion is amusing. * The right stuff to use is called "electrical contract cleaner", available from Radio Shack or equivalent. It will even revive keyboards after a user dumps a can worth of soda into them. Alcohol is a decent second-choice alternative. -- -Chuck [*]: "Amusing" since this entire reply was prompted by a memory of someone who tried to fix scratched audio CD's using toothpaste...which sometimes even worked, but more often generated some remarkable skipping remixes that would confuse CD players. :-) I'd imagine you could make a good markup reselling ski wax, if you started a rumor on overclocking sites that waxing your RAM would make your computer go faster ("hey! this stuff makes skis and snowboards go faster, right?") Contrary to popular opinion, however, computer memory does not run faster when you apply wax to it. In other words, no car polish. _______________________________________________ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscribe@freebsd.org" From owner-freebsd-performance@FreeBSD.ORG Thu Dec 9 17:49:07 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 1EE0B16A4CE for ; Thu, 9 Dec 2004 17:49:07 +0000 (GMT) Received: from adsl-63-198-35-122.dsl.snfc21.pacbell.net (adsl-63-198-35-122.dsl.snfc21.pacbell.net [63.198.35.122]) by mx1.FreeBSD.org (Postfix) with ESMTP id C4C5843D53 for ; Thu, 9 Dec 2004 17:49:06 +0000 (GMT) (envelope-from jin@george.lbl.gov) Received: from [IPv6:::1] (localhost.pacbell.net [IPv6:::1]) ESMTP id iB9Hn5jl000478; Thu, 9 Dec 2004 09:49:05 -0800 (PST) (envelope-from jin@george.lbl.gov) Message-ID: <41B89011.3040700@george.lbl.gov> Date: Thu, 09 Dec 2004 09:49:05 -0800 From: "Jin Guojun [VFFS]" User-Agent: Mozilla/5.0 (X11; U; FreeBSD i386; en-US; rv:1.7.2) Gecko/20040912 X-Accept-Language: zh, zh-CN, en MIME-Version: 1.0 To: Mike Horwath References: <20041209170655.GH45512@octanews.net> In-Reply-To: <20041209170655.GH45512@octanews.net> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit X-Mailman-Approved-At: Thu, 09 Dec 2004 19:25:57 +0000 cc: freebsd-performance@freebsd.org cc: Bill Anderson Subject: Re: 2120S poor performance X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 09 Dec 2004 17:49:07 -0000 Mike Horwath wrote: >On Wed, Dec 08, 2004 at 11:51:21AM -0500, Bill Anderson wrote: > > >>I've got a 3x72g RAID5 array with U320 disks on an Adaptec 2120S >>controller under Freebsd 4.10. I'm getting about 25MB/s for sequential >>reads/writes (e.g. dd if=/dev/aacd0s1b of=/dev/null bs=32k). I've >>turned on write caching for the container (container set cache >>/write_cache_enable) and turned off the read caching, per previous posts >>to this list. I'm going to try setting hw.aac.iosize_max to 96k tonight >>to see if that helps. I'm currently using an IDE disk that gets about >>45MB/s, so I can hardly consider it an "upgrade" to switch to a 25MB/s >>U320 SCSI system. :/ >> >>I've also read that the 2120S is slow because of its design. I'm trying >>to figure out whether the performance can be increased significantly (I >>saw a posting of a linux user getting 37.5MB/s, which although still slow >>might be acceptable), or if I'm better off getting a new card. I'm >>thinking the latter is the case based on previous postings, but since it >>will probably be quite expensive to replace it, I wanted to get some more >>data. >> >>What performance should I be expecting from a decent U320 RAID5 >>controller? >> >>Has anyone gotten a 2120S to perform above 30Mb/s in FreeBSD? >> >>What's the cheapest controller that still gives reasonable performance? >>(If you could give a couple different ones, with their associated >>performance (under FreeBSD), or tell me where to find such information, >>that would be great) >> >> > >I know I left quite a bit of cruft above, but I don't see these kinds >of performance numbers you are seeing. > >First, SCSI does a lot with overlapping I/O and tagged queuing, only >the newest IDE (SATA) systems are starting to support such things. > >Second, SCSI disks run at a faster RPM most of the time (only the WD >Raptor hits the 10K mark), lowering seek time significantly. > >Third, my numbers: > >31MB/sec using > > dd if=/dev/aacd0s1b of=/dev/null bs=32k count=20000 > >34MB/sec using > > dd if=/dev/aacd0s1b of=/dev/null bs=64k count=20000 > >33MB/sec using > > dd if=/dev/aacd0s1b of=/dev/null bs=128k count=20000 > I do not remember where the original problem of 2120s was posted. The read in RAID 5 mode is 40-56 MB/s, where write rate is 5-13 MB/s. The most problem is at writing to RAID, average around 8 MB/s. SCSI buses self is Okay because doing multiple R/W without RAID can get maximum disk I/O. Soft RAID, like Vinum, also slows down disk array performance. SO, it sounds like that RAID design has big overhead to kill the performance. dd if=/dev/aacd0s1b of=/dev/null bs=128k count=20000 dev/null bs=128k count=20000 20000+0 records in 20000+0 records out 2621440000 bytes transferred in 61.651078 secs (42520587 bytes/sec) Adaptec guy told me that this is a design issue. One older model, which may be out of market, is performing better, but I would not expect a large difference as soft RAID do not give better performance. -- ------------ Jin Guojun ----------- v --- jin@george.lbl.gov --- Distributed Systems Department http://www.dsd.lbl.gov/~jin Lawrence Berkeley National Laboratory, Berkeley, CA 94720 From owner-freebsd-performance@FreeBSD.ORG Thu Dec 9 19:52:41 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 4F8B016A4CE for ; Thu, 9 Dec 2004 19:52:41 +0000 (GMT) Received: from stewie.obfuscated.net (stewie.obfuscated.net [66.118.188.125]) by mx1.FreeBSD.org (Postfix) with ESMTP id 0D01A43D39 for ; Thu, 9 Dec 2004 19:52:41 +0000 (GMT) (envelope-from meconlen@obfuscated.net) Received: from [192.168.1.101] (653259hfc120.tampabay.rr.com [65.32.59.120]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by stewie.obfuscated.net (Postfix) with ESMTP id 3223A6153 for ; Thu, 9 Dec 2004 14:52:40 -0500 (EST) Mime-Version: 1.0 (Apple Message framework v619) In-Reply-To: <41B89011.3040700@george.lbl.gov> References: <20041209170655.GH45512@octanews.net> <41B89011.3040700@george.lbl.gov> Content-Type: text/plain; charset=US-ASCII; format=flowed Message-Id: Content-Transfer-Encoding: 7bit Cc: freebsd-performance@freebsd.org From: Michael E.Conlen Date: Thu, 9 Dec 2004 14:51:49 -0500 X-Mailer: Apple Mail (2.619) Subject: Re: 2120S poor performance X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 09 Dec 2004 19:52:41 -0000 On Dec 9, 2004, at 12:49 PM, Jin Guojun [VFFS] wrote: > I do not remember where the original problem of 2120s was posted. > The read in RAID 5 mode is 40-56 MB/s, where write rate is 5-13 MB/s. > The most problem is at writing to RAID, average around 8 MB/s. > SCSI buses self is Okay because doing multiple R/W without RAID > can get maximum disk I/O. Soft RAID, like Vinum, also slows down > disk array performance. SO, it sounds like that RAID design has > big overhead to kill the performance. > > dd if=/dev/aacd0s1b of=/dev/null bs=128k count=20000 > dev/null bs=128k count=20000 > 20000+0 records in > 20000+0 records out > 2621440000 bytes transferred in 61.651078 secs (42520587 bytes/sec) > > Adaptec guy told me that this is a design issue. One older model, > which may be out of market, is performing better, but I would not > expect a large difference as soft RAID do not give better performance. Poorly implemented RAID-5 will often times incur a read operation for every write so that it can calculate the checksum value. Some implementations will identify when an entire stripe is being written and avoid it when possible. Try tuning the FS so that the blocks in the FS match up with the stripe size in the controller. I believe this will force every write to be a stripe. -- Michael Conlen meconlen@obfuscated.net From owner-freebsd-performance@FreeBSD.ORG Fri Dec 10 00:16:56 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id C810C16A4CE for ; Fri, 10 Dec 2004 00:16:56 +0000 (GMT) Received: from mail2.dbitech.ca (radius.wavefire.com [64.141.13.252]) by mx1.FreeBSD.org (Postfix) with SMTP id 0CB5843D2F for ; Fri, 10 Dec 2004 00:16:56 +0000 (GMT) (envelope-from darcy@wavefire.com) Received: (qmail 4298 invoked from network); 10 Dec 2004 01:24:10 -0000 Received: from dbitech.wavefire.com (HELO ?64.141.15.253?) (darcy@64.141.15.253) by radius.wavefire.com with SMTP; 10 Dec 2004 01:24:10 -0000 From: Darcy Buskermolen Organization: Wavefire Technologies Corp. To: freebsd-performance@freebsd.org Date: Thu, 9 Dec 2004 16:16:53 -0800 User-Agent: KMail/1.6.2 References: <200412090049.iB90ntMx029318@mail.cs.ait.ac.th> <200412100041.46272.akhthar@carmatec.com> In-Reply-To: <200412100041.46272.akhthar@carmatec.com> MIME-Version: 1.0 Content-Disposition: inline Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Message-Id: <200412091616.53608.darcy@wavefire.com> Subject: Re: Memory problems:-( X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 10 Dec 2004 00:16:56 -0000 On December 9, 2004 11:11 am, Akhthar Parvez. K wrote: > > This is not a bad thought, although simply removing the RAM and reinserting > it properly is often enough to reset the memory and wipe the contacts clean > of any minor oxidation on the contacts. > > Toothpaste or other abrasives are excessive, although the notion is > amusing. * > > The right stuff to use is called "electrical contract cleaner", available > from Radio Shack or equivalent. It will even revive keyboards after a user > dumps a can worth of soda into them. > > Alcohol is a decent second-choice alternative. If you are using Alcohol, make sure that it is 99.9% pure isoproponal (rubbing alcohol) and not 70%. The 70% leaves a residue behind that defeats the purpose of cleaning to beging with. -- Darcy Buskermolen Wavefire Technologies Corp. ph: 250.717.0200 fx: 250.763.1759 http://www.wavefire.com From owner-freebsd-performance@FreeBSD.ORG Sat Dec 11 01:08:54 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 7893F16A4CE for ; Sat, 11 Dec 2004 01:08:54 +0000 (GMT) Received: from dragon.nuxi.com (trang.nuxi.com [66.93.134.19]) by mx1.FreeBSD.org (Postfix) with ESMTP id 4D58743D2F for ; Sat, 11 Dec 2004 01:08:54 +0000 (GMT) (envelope-from obrien@NUXI.com) Received: from dragon.nuxi.com (obrien@localhost [127.0.0.1]) by dragon.nuxi.com (8.13.1/8.13.1) with ESMTP id iBB18nnJ052257; Fri, 10 Dec 2004 17:08:49 -0800 (PST) (envelope-from obrien@dragon.nuxi.com) Received: (from obrien@localhost) by dragon.nuxi.com (8.13.1/8.13.1/Submit) id iBB18nLT052249; Fri, 10 Dec 2004 17:08:49 -0800 (PST) (envelope-from obrien) Date: Fri, 10 Dec 2004 17:08:49 -0800 From: "David O'Brien" To: Jon Noack Message-ID: <20041211010849.GE50516@dragon.nuxi.com> References: <17418.69.53.57.66.1102618255.squirrel@69.53.57.66> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <17418.69.53.57.66.1102618255.squirrel@69.53.57.66> User-Agent: Mutt/1.4.1i X-Operating-System: FreeBSD 6.0-CURRENT Organization: The NUXI BSD Group X-Pgp-Rsa-Fingerprint: B7 4D 3E E9 11 39 5F A3 90 76 5D 69 58 D9 98 7A X-Pgp-Rsa-Keyid: 1024/34F9F9D5 cc: freebsd-performance@freebsd.org cc: Bill Anderson Subject: Re: 2120S poor performance X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list Reply-To: freebsd-performance@freebsd.org List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 11 Dec 2004 01:08:54 -0000 On Thu, Dec 09, 2004 at 12:50:55PM -0600, Jon Noack wrote: > Bill Anderson wrote: > > I've got a 3x72g RAID5 array with U320 disks on an Adaptec 2120S > > controller under Freebsd 4.10. I'm getting about 25MB/s for sequential .. > > What's the cheapest controller that still gives reasonable performance? > > (If you could give a couple different ones, with their associated > > performance (under FreeBSD), or tell me where to find such information, > > that would be great) > > The asr driver is in need of a lot of attention. .. > There are many quality, (relatively) cheap controllers out there, but if > you're looking to stick with Adaptec get a 2130S. The 2130S uses the aac > driver, which is quite well tuned and maintained by Scott Long. The Adaptec SCSI RAID 2120S is not an I2O(asr) controller -- it is an aac(4) controller. The 'out of market' controller mentioned in this thread is the work horse 5400S. -- -- David (obrien@FreeBSD.org) From owner-freebsd-performance@FreeBSD.ORG Sat Dec 11 01:59:55 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 6FF5F16A4CE for ; Sat, 11 Dec 2004 01:59:55 +0000 (GMT) Received: from multiplay.co.uk (www1.multiplay.co.uk [212.42.16.7]) by mx1.FreeBSD.org (Postfix) with ESMTP id 6579843D5D; Sat, 11 Dec 2004 01:59:54 +0000 (GMT) (envelope-from killing@multiplay.co.uk) Received: from vader ([212.135.219.179]) by multiplay.co.uk (multiplay.co.uk [212.42.16.7]) (MDaemon.PRO.v7.2.1.R) with ESMTP id md50000786530.msg; Sat, 11 Dec 2004 01:51:27 +0000 Message-ID: <010b01c4df25$14af3ff0$b3db87d4@multiplay.co.uk> From: "Steven Hartland" To: , "Jon Noack" References: <17418.69.53.57.66.1102618255.squirrel@69.53.57.66> <20041211010849.GE50516@dragon.nuxi.com> Date: Sat, 11 Dec 2004 01:59:39 -0000 MIME-Version: 1.0 Content-Type: text/plain; format=flowed; charset="iso-8859-1"; reply-type=original Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2900.2180 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.2180 X-Spam-Processed: multiplay.co.uk, Sat, 11 Dec 2004 01:51:27 +0000 (not processed: message from valid local sender) X-MDRemoteIP: 212.135.219.179 X-Return-Path: killing@multiplay.co.uk X-MDAV-Processed: multiplay.co.uk, Sat, 11 Dec 2004 01:51:28 +0000 cc: freebsd-performance@freebsd.org cc: Bill Anderson Subject: Re: 2120S poor performance X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 11 Dec 2004 01:59:55 -0000 2015S is though which is what I tested and it gives very similar poor results dmesg |grep asr: asr0: ADAPTEC 2015S FW Rev. 3B05, 2 channel, 256 CCBs, Protocol I2O Does anyone have any RAID5 SCSI card installed that does give good results in a straight dd test? Steve >> The asr driver is in need of a lot of attention. > .. >> There are many quality, (relatively) cheap controllers out there, but if >> you're looking to stick with Adaptec get a 2130S. The 2130S uses the aac >> driver, which is quite well tuned and maintained by Scott Long. > > The Adaptec SCSI RAID 2120S is not an I2O(asr) controller -- it is an > aac(4) controller. The 'out of market' controller mentioned in this > thread is the work horse 5400S. ================================================ This e.mail is private and confidential between Multiplay (UK) Ltd. and the person or entity to whom it is addressed. In the event of misdirection, the recipient is prohibited from using, copying, printing or otherwise disseminating it or any information contained in it. In the event of misdirection, illegible or incomplete transmission please telephone (023) 8024 3137 or return the E.mail to postmaster@multiplay.co.uk. From owner-freebsd-performance@FreeBSD.ORG Sat Dec 11 03:41:16 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id C078516A4CE for ; Sat, 11 Dec 2004 03:41:16 +0000 (GMT) Received: from smtp101.rog.mail.re2.yahoo.com (smtp101.rog.mail.re2.yahoo.com [206.190.36.79]) by mx1.FreeBSD.org (Postfix) with SMTP id 417D943D48 for ; Sat, 11 Dec 2004 03:41:16 +0000 (GMT) (envelope-from mikej@rogers.com) Received: from unknown (HELO 172.16.0.1) (mikej@69.193.222.195 with login) by smtp101.rog.mail.re2.yahoo.com with SMTP; 11 Dec 2004 03:41:15 -0000 Received: from 172.16.0.200 (SquirrelMail authenticated user mikej); by 172.16.0.1 with HTTP; Fri, 10 Dec 2004 22:41:05 -0500 (EST) Message-ID: <3304.172.16.0.200.1102736465.squirrel@172.16.0.200> In-Reply-To: <010b01c4df25$14af3ff0$b3db87d4@multiplay.co.uk> References: <17418.69.53.57.66.1102618255.squirrel@69.53.57.66> <20041211010849.GE50516@dragon.nuxi.com> <010b01c4df25$14af3ff0$b3db87d4@multiplay.co.uk> Date: Fri, 10 Dec 2004 22:41:05 -0500 (EST) From: "Mike Jakubik" To: "Steven Hartland" User-Agent: SquirrelMail/1.4.3a X-Mailer: SquirrelMail/1.4.3a MIME-Version: 1.0 Content-Type: text/plain;charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Priority: 3 (Normal) Importance: Normal cc: freebsd-performance@freebsd.org cc: Bill Anderson Subject: Re: 2120S poor performance X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 11 Dec 2004 03:41:16 -0000 Steven Hartland said: > 2015S is though which is what I tested and it gives very similar poor > results > dmesg |grep asr: > asr0: ADAPTEC 2015S FW Rev. 3B05, 2 channel, 256 CCBs, Protocol I2O > > Does anyone have any RAID5 SCSI card installed that does give good > results in a straight dd test? This is what i get on the following (FreeBSD 5.3-STABLE #0: Sun Nov 7 03:25:47 EST 2004): --- da0 at asr0 bus 0 target 0 lun 0 da0: Fixed Direct Access SCSI-2 device da0: Tagged Queueing Enabled da0: 70006MB (143372288 512 byte sectors: 255H 63S/T 8924C) # diskinfo -tv da0 da0 512 # sectorsize 73406611456 # mediasize in bytes (68G) 143372288 # mediasize in sectors 8924 # Cylinders according to firmware. 255 # Heads according to firmware. 63 # Sectors according to firmware. Seek times: Full stroke: 250 iter in 1.222343 sec = 4.889 msec Half stroke: 250 iter in 1.189145 sec = 4.757 msec Quarter stroke: 500 iter in 2.315375 sec = 4.631 msec Short forward: 400 iter in 1.821531 sec = 4.554 msec Short backward: 400 iter in 1.936996 sec = 4.842 msec Seq outer: 2048 iter in 2.296997 sec = 1.122 msec Seq inner: 2048 iter in 1.991383 sec = 0.972 msec Transfer rates: outside: 102400 kbytes in 3.810755 sec = 26871 kbytes/sec middle: 102400 kbytes in 3.907428 sec = 26206 kbytes/sec inside: 102400 kbytes in 3.235306 sec = 31651 kbytes/sec --- I'm dissapointed in the transfer rates.. seek times however are nice. From owner-freebsd-performance@FreeBSD.ORG Sat Dec 11 16:51:43 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 0EF9816A4CE for ; Sat, 11 Dec 2004 16:51:43 +0000 (GMT) Received: from tog.net (tog.net [216.89.226.5]) by mx1.FreeBSD.org (Postfix) with ESMTP id AC52B43D39 for ; Sat, 11 Dec 2004 16:51:42 +0000 (GMT) (envelope-from ender@tog.net) Received: by tog.net (Postfix, from userid 96) id 4373929B76A; Sat, 11 Dec 2004 11:51:42 -0500 (EST) Received: from tog.net (host-216-89-225-139.terranova.net [216.89.225.139]) by tog.net (Postfix) with ESMTP id 5620C29B746 for ; Sat, 11 Dec 2004 11:51:40 -0500 (EST) Message-ID: <41BB2699.7090604@tog.net> Date: Sat, 11 Dec 2004 11:55:53 -0500 From: Ender User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.6b) Gecko/20031205 Thunderbird/0.4 X-Accept-Language: en-us, en MIME-Version: 1.0 To: freebsd-performance@freebsd.org Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Spam-Checker-Version: SpamAssassin 3.0.1-terranovanet_v6 (2004-10-22) on doppelganger.terranova.net X-Spam-Level: X-Spam-Status: No, score=-2.4 required=7.1 tests=ALL_TRUSTED,AWL,BAYES_50 autolearn=ham version=3.0.1-terranovanet_v6 Subject: Device IDT eating all cpu X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 11 Dec 2004 16:51:43 -0000 FreeBSD 5.3-BETA7 #6 device idt spams my console 0 with the following message: idt0: i= 0, status=64320000 This eats up all my cpu, and really lags the system. Everything else about the device is working properly. (passing network traffic, ect) If there is anymore infomation i can provide to help, just let me know idt0: port 0x9400-0x94ff mem 0xe3004000-0xe3004fff irq 5 at device 6.0 on pci1 idt0: [GIANT-LOCKED] idt0: ATM card is Fore LE25, PHY=77105 idt0: 32K words of RAM idt0: MAC address 00:20:48:21:11:4e, HWrev=2 idt0: i= 0, status=64320000 idt0: i= 0, status=64320000 idt0: i= 0, status=64320000 idt0: i= 0, status=64320000