From owner-freebsd-performance@FreeBSD.ORG Sun Apr 15 00:49:31 2007 Return-Path: X-Original-To: freebsd-performance@freebsd.org Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 307AD16A400 for ; Sun, 15 Apr 2007 00:49:31 +0000 (UTC) (envelope-from unixtools@hotmail.com) Received: from bay0-omc1-s31.bay0.hotmail.com (bay0-omc1-s31.bay0.hotmail.com [65.54.246.103]) by mx1.freebsd.org (Postfix) with ESMTP id 1CA4F13C44B for ; Sun, 15 Apr 2007 00:49:31 +0000 (UTC) (envelope-from unixtools@hotmail.com) Received: from hotmail.com ([65.54.250.39]) by bay0-omc1-s31.bay0.hotmail.com with Microsoft SMTPSVC(6.0.3790.2668); Sat, 14 Apr 2007 17:37:31 -0700 Received: from mail pickup service by hotmail.com with Microsoft SMTPSVC; Sat, 14 Apr 2007 17:37:31 -0700 Message-ID: Received: from 65.54.250.200 by by115fd.bay115.hotmail.msn.com with HTTP; Sun, 15 Apr 2007 00:37:27 GMT X-Originating-IP: [24.228.29.56] X-Originating-Email: [unixtools@hotmail.com] X-Sender: unixtools@hotmail.com In-Reply-To: <20070330122858.93392f43.teklimbu@wlink.com.np> From: "Sunil Sunder Raj" To: teklimbu@wlink.com.np, freebsd-performance@freebsd.org Date: Sun, 15 Apr 2007 00:37:27 +0000 Mime-Version: 1.0 Content-Type: text/plain; format=flowed X-OriginalArrivalTime: 15 Apr 2007 00:37:31.0130 (UTC) FILETIME=[406D2DA0:01C77EF6] Cc: Subject: RE: FreeBSD-6.1 reboots due to high mbufs 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: Sun, 15 Apr 2007 00:49:31 -0000 Hi, Seems to be an old posting. Were you able to solve this problem. What is your ifconfig output. Sunil Sunder Raj http://daemon.in >From: Tek Bahadur Limbu >To: freebsd-performance@freebsd.org >Subject: FreeBSD-6.1 reboots due to high mbufs >Date: Fri, 30 Mar 2007 12:28:58 +0545 > >-----BEGIN PGP SIGNED MESSAGE----- >Hash: SHA1 > >Dear All, > >I am quite new to FreeBSD in general and I also don't have much experience >and knowledge as required by this mailing list. I had posted this on >freebsd-questions but till now have not received any kind of feedback which >is fine. > >I need some help and suggestions from this mailing list if permissible. I >am sorry if I am posting this to the wrong mailing list. > >I have squid proxy server running on a FreeBSD-6.1 (amd) box. I have been >facing this problem for sometime now. It's related to mbufs. For some >reasons, my mbufs usage is extremely high. This high mbufs usage causes >slow responses from Squid and in rare occasions, it even causes my server >to reboot or locks out SSH sessions. > >This FreeBSD-6.1 squid box serves about 3000-4000 users. > >Server specs are: > >Dell 430 Intel(R) Pentium(R) D CPU 3.00GHz-Dual CPU >real memory = 2145959936 (2046 MB) >bge0: traffic load: 10Mb/s (through satellite) > >root@server# netstat -m > >66713/2677/69390 mbufs in use (current/cache/total) >66686/2210/68896/0 mbuf clusters in use (current/cache/total/max) >66686/2178 mbuf+clusters out of packet secondary zone in use >(current/cache) >0/0/0/0 4k (page size) jumbo clusters in use (current/cache/total/max) >0/0/0/0 9k jumbo clusters in use (current/cache/total/max) >0/0/0/0 16k jumbo clusters in use (current/cache/total/max) >150050K/5089K/155139K bytes allocated to network (current/cache/total) >0/0/0 requests for mbufs denied (mbufs/clusters/mbuf+clusters) >0/0/0 requests for jumbo clusters denied (4k/9k/16k) >0/0/0 sfbufs in use (current/peak/max) >0 requests for sfbufs denied >0 requests for sfbufs delayed >0 requests for I/O initiated by sendfile >0 calls to protocol drain routines > > >/etc/sysctl.conf: > >kern.maxfilesperproc=8192 >kern.ipc.somaxconn=8192 >kern.maxprocperuid=8192 >net.inet.ip.portrange.hifirst=10000 >net.inet.ip.portrange.first=30000 >net.inet.ip.portrange.last=65535 >net.inet.tcp.recvspace=65535 >net.inet.tcp.sendspace=65535 >net.inet.tcp.blackhole=2 >net.inet.udp.blackhole=1 >net.inet.ip.fw.verbose_limit=5000 >net.inet.icmp.icmplim=0 >net.inet.tcp.msl=3000 >net.isr.direct=1 >vfs.read_max=16 > >/boot/loader.conf: > >kern.ipc.nmbclusters=0 >kern.maxusers=0 >kern.maxfiles=16384 >kern.maxproc=8192 > >Relevant Kernel Options: > >options IPFIREWALL >options IPFIREWALL_VERBOSE >options IPFIREWALL_FORWARD >options IPFIREWALL_VERBOSE_LIMIT=5000 >options IPFIREWALL_FORWARD_EXTENDED > >options MSGMNB=16384 >options MSGMNI=41 >options MSGSEG=2049 >options MSGSSZ=64 >options MSGTQL=2048 >options SHMSEG=128 >options SHMMNI=192 >options SHMMAX=33554432 >options SHMMIN=1 >options SHMALL=8192 > >options HZ=1000 #Polling Enabled > > >Note: Running IPFW. Also my other FreeBSD-4.x servers don't seem to suffer >from this problem. They have almost the same IPFW, Squid and Sysctl >configs. > >Any feedback and suggestions will be highly appreciated. > >Thanking you... > > >- -- > > >With best regards and good wishes, > >Yours sincerely, > >Tek Bahadur Limbu > >(TAG/TDG Group) >Jwl Systems Department > >Worldlink Communications Pvt. Ltd. > >Jawalakhel, Nepal > >http://www.wlink.com.np >-----BEGIN PGP SIGNATURE----- >Version: GnuPG v1.4.2.2 (FreeBSD) > >iD8DBQFGDLGuVrOl+eVhOvYRAk8dAJ9dumhWCNTp6iHUy5PsQ0ie2OdMqQCgjG8u >eTxXgRCNSe3R2mmTGSbuSqo= >=mVGh >-----END PGP SIGNATURE----- >_______________________________________________ >freebsd-performance@freebsd.org mailing list >http://lists.freebsd.org/mailman/listinfo/freebsd-performance >To unsubscribe, send any mail to >"freebsd-performance-unsubscribe@freebsd.org" _________________________________________________________________ Interest Rates Fall Again! $430,000 Mortgage for $1,399/mo - Calculate new payment http://www.lowermybills.com/lre/index.jsp?sourceid=lmb-9632-18679&moid=7581 From owner-freebsd-performance@FreeBSD.ORG Sun Apr 15 19:13:53 2007 Return-Path: X-Original-To: freebsd-performance@freebsd.org Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 65B2216A57B for ; Sun, 15 Apr 2007 19:13:53 +0000 (UTC) (envelope-from daved@tamu.edu) Received: from sr-7-int.cis.tamu.edu (smtp-relay.TAMU.EDU [165.91.22.120]) by mx1.freebsd.org (Postfix) with ESMTP id 37B8113C457 for ; Sun, 15 Apr 2007 19:13:53 +0000 (UTC) (envelope-from daved@tamu.edu) Received: from localhost (localhost.tamu.edu [127.0.0.1]) by sr-7-int.cis.tamu.edu (Postfix) with ESMTP id D17EB4F77B for ; Sun, 15 Apr 2007 13:58:04 -0500 (CDT) Received: from [10.0.1.2] (pool-71-126-195-96.herntx.dsl-w.verizon.net [71.126.195.96]) (using TLSv1 with cipher AES128-SHA (128/128 bits)) (No client certificate requested) by sr-7-int.cis.tamu.edu (Postfix) with ESMTP id 3B1044F6A8 for ; Sun, 15 Apr 2007 13:58:04 -0500 (CDT) Mime-Version: 1.0 (Apple Message framework v752.2) To: freebsd-performance@freebsd.org Message-Id: <6529290C-6D7B-4CCB-94F5-BB44AD999848@tamu.edu> Content-Type: multipart/signed; micalg=sha1; boundary=Apple-Mail-3-787787377; protocol="application/pkcs7-signature" From: David Duchscher Date: Sun, 15 Apr 2007 13:58:02 -0500 X-Mailer: Apple Mail (2.752.2) X-Virus-Scanned: amavisd-new at tamu.edu X-Content-Filtered-By: Mailman/MimeDel 2.1.5 Subject: 10 gigabit performance 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: Sun, 15 Apr 2007 19:13:53 -0000 --Apple-Mail-3-787787377 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Anybody have any performance numbers on 10 gigabit interfaces? We are looking for numbers on both routing and bridging for the various packet sizes. We would like to see where FreeBSD sits with other products we are comparing and do not have any 10 gigabit interfaces to test as yet. On that note, if anybody has any recommendations or comments on 10 gigabit network cards that work with freebsd, it would be greatly appreciated. Thanks for your time, -- DaveD --Apple-Mail-3-787787377-- From owner-freebsd-performance@FreeBSD.ORG Mon Apr 16 00:45:17 2007 Return-Path: X-Original-To: freebsd-performance@freebsd.org Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id C65A516A400 for ; Mon, 16 Apr 2007 00:45:17 +0000 (UTC) (envelope-from markir@paradise.net.nz) Received: from smtp3.clear.net.nz (smtp3.clear.net.nz [203.97.33.64]) by mx1.freebsd.org (Postfix) with ESMTP id 914E613C455 for ; Mon, 16 Apr 2007 00:45:17 +0000 (UTC) (envelope-from markir@paradise.net.nz) Received: from [192.168.1.11] (121-72-69-46.dsl.telstraclear.net [121.72.69.46]) by smtp3.clear.net.nz (CLEAR Net Mail) with ESMTP id <0JGK00NK2ENS3L10@smtp3.clear.net.nz> for freebsd-performance@freebsd.org; Mon, 16 Apr 2007 12:28:41 +1200 (NZST) Date: Mon, 16 Apr 2007 12:28:33 +1200 From: Mark Kirkwood In-reply-to: <20070323181437.GA94251@deeboz.ca> To: Sally Janghos Message-id: <4622C331.3080800@paradise.net.nz> MIME-version: 1.0 Content-type: text/plain; charset=ISO-8859-1; format=flowed Content-transfer-encoding: 7bit References: <006101c76cda$313892d0$da11e00a@Seelye> <20070323181437.GA94251@deeboz.ca> User-Agent: Thunderbird 1.5.0.10 (X11/20070313) Cc: Aaron Seelye , freebsd-performance@freebsd.org Subject: Re: Where to troubleshoot Intel PRO/1000 performance problems? 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: Mon, 16 Apr 2007 00:45:17 -0000 Sally Janghos wrote: > Aaron, > > Thanks for your reply. Here is some output from some dd's on the disk that I'm reading/writing to. > >> dd if=/dev/zero of=testfile count=1000000;dd if=testfile of=/dev/null > 1000000+0 records in > 1000000+0 records out > 512000000 bytes transferred in 27.951769 secs (18317267 bytes/sec) > 1000000+0 records in > 1000000+0 records out > 512000000 bytes transferred in 16.864945 secs (30358830 bytes/sec) >> dd if=/dev/zero of=testfile count=1000000 ; dd if=testfile of=/dev/null > 1000000+0 records in > 1000000+0 records out > 512000000 bytes transferred in 28.492921 secs (17969376 bytes/sec) > 1000000+0 records in > 1000000+0 records out > 512000000 bytes transferred in 16.605797 secs (30832607 bytes/sec) > > Am I reading right, the max network transfer rate from this disk will be between 143Mbit and 246Mbit/sec? Is there a way to determine if there is an IRQ conflict? How do you find out what IRQ's are currently used by the system? To see bigger transfer rates you might want to retest with a bigger block size for the dd - e.g (FreeBSD 6.2 on a PIII 1.26Ghz with Promise ATA controller 2xSeagate IDE RAID0): $ dd if=/dev/zero of=testfile bs=8k count=64000 ; dd if=testfile bs=8k of=/dev/null 64000+0 records in 64000+0 records out 524288000 bytes transferred in 11.624216 secs (45103084 bytes/sec) 64000+0 records in 64000+0 records out 524288000 bytes transferred in 1.544103 secs (339542136 bytes/sec) Note for the read is clearly coming from memory (2 IDE disks will do 100 MBytes/s max - not 300!). Cheers Mark From owner-freebsd-performance@FreeBSD.ORG Mon Apr 16 01:50:22 2007 Return-Path: X-Original-To: freebsd-performance@freebsd.org Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 207B416A401 for ; Mon, 16 Apr 2007 01:50:22 +0000 (UTC) (envelope-from killing@multiplay.co.uk) Received: from multiplay.co.uk (core6.multiplay.co.uk [85.236.96.23]) by mx1.freebsd.org (Postfix) with ESMTP id AF67913C45B for ; Mon, 16 Apr 2007 01:50:21 +0000 (UTC) (envelope-from killing@multiplay.co.uk) X-Spam-Checker-Version: SpamAssassin 3.1.5 (2006-08-29) on core6.multiplay.co.uk X-Spam-Level: X-Spam-Status: No, score=-24.7 required=6.0 tests=BAYES_00, USER_IN_WHITELIST, USER_IN_WHITELIST_TO autolearn=ham version=3.1.5 Received: from r2d2 ([212.135.219.182]) by multiplay.co.uk (multiplay.co.uk [85.236.96.23]) (MDaemon PRO v9.5.4) with ESMTP id md50003736584.msg for ; Mon, 16 Apr 2007 02:34:46 +0100 Message-ID: <008801c77fc7$74ad6a90$b6db87d4@multiplay.co.uk> From: "Steven Hartland" To: Date: Mon, 16 Apr 2007 02:35:02 +0100 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 6.00.2900.3028 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.3028 X-MDRemoteIP: 212.135.219.182 X-Return-Path: killing@multiplay.co.uk X-Envelope-From: killing@multiplay.co.uk X-MDaemon-Deliver-To: freebsd-performance@freebsd.org X-Spam-Processed: multiplay.co.uk, Mon, 16 Apr 2007 02:34:47 +0100 X-MDAV-Processed: multiplay.co.uk, Mon, 16 Apr 2007 02:34:47 +0100 Subject: Great ftp performance with 6.2-RELEASE 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: Mon, 16 Apr 2007 01:50:22 -0000 Just a quick note to let everyone know the outstanding performance we achieved using 6.2-RELEASE on bge fibre gig via an Extreme Black Diamond and base ftp + proftpd. When transferring a single ISO image via the above setup we see 92MB/s. Both machines where AMD Opteron based with HighPoint 1820a RAID 5 array's. The only real change from stock to achieve this was to disable inflight: net.inet.tcp.inflight.enable=0 Good work guys. Regards Steve ================================================ 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 +44 845 868 1337 or return the E.mail to postmaster@multiplay.co.uk. From owner-freebsd-performance@FreeBSD.ORG Tue Apr 17 10:15:35 2007 Return-Path: X-Original-To: freebsd-performance@freebsd.org Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 9046116A404 for ; Tue, 17 Apr 2007 10:15:35 +0000 (UTC) (envelope-from teklimbu@wlink.com.np) Received: from smtp5.wlink.com.np (smtp5.wlink.com.np [202.79.32.52]) by mx1.freebsd.org (Postfix) with SMTP id 5763213C459 for ; Tue, 17 Apr 2007 10:15:28 +0000 (UTC) (envelope-from teklimbu@wlink.com.np) Received: (qmail 26524 invoked from network); 17 Apr 2007 10:15:20 -0000 Received: from unknown (HELO smtp8.wlink.com.np) (202.79.32.38) by 0 with SMTP; 17 Apr 2007 10:15:20 -0000 Received: (qmail 37049 invoked by uid 98); 17 Apr 2007 10:15:20 -0000 Received: from 202.79.32.77 by smtp8.wlink.com.np (envelope-from , uid 1004) with qmail-scanner-1.25 (clamdscan: 0.88.4/1879. Clear:RC:1(202.79.32.77):. Processed in 0.074221 secs); 17 Apr 2007 10:15:20 -0000 X-Qmail-Scanner-Mail-From: teklimbu@wlink.com.np via smtp8.wlink.com.np X-Qmail-Scanner: 1.25 (Clear:RC:1(202.79.32.77):. Processed in 0.074221 secs) Received: from smtp2.wlink.com.np (202.79.32.77) by smtp8.wlink.com.np with SMTP; 17 Apr 2007 10:15:20 -0000 Received: (qmail 20192 invoked by uid 516); 17 Apr 2007 10:15:19 -0000 Received: from [202.79.36.216] (HELO teklimbu.wlink.com.np) by smtp2.wlink.com.np (qmail-smtpd) with SMTP; 17 Apr 2007 10:15:14 -0000 (Tue, 17 Apr 2007 16:00:14 +0545) Received: from teklimbu.wlink.com.np (teklimbu.wlink.com.np [202.79.36.216]) by teklimbu.wlink.com.np (Postfix) with SMTP id E467D73032; Tue, 17 Apr 2007 16:00:10 +0545 (NPT) Date: Tue, 17 Apr 2007 16:00:10 +0545 From: Tek Bahadur Limbu To: "Sunil Sunder Raj" Message-Id: <20070417160010.afa7b31f.teklimbu@wlink.com.np> In-Reply-To: References: <20070330122858.93392f43.teklimbu@wlink.com.np> Organization: Worldlink Communications Pvt. Ltd. X-Mailer: Sylpheed version 2.2.0 (GTK+ 2.8.12; i386-portbld-freebsd6.1) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Spam-Check-By: smtp2.wlink.com.np Spam: No ; 0.1 / 7.0 X-Spam-Status-WL: No, hits=0.1 required=7.0 Cc: freebsd-performance@freebsd.org Subject: Re: FreeBSD-6.1 reboots due to high mbufs 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: Tue, 17 Apr 2007 10:15:35 -0000 -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Sun, 15 Apr 2007 00:37:27 +0000 "Sunil Sunder Raj" wrote: > Hi, > > Seems to be an old posting. Were you able to solve this problem. What is > your ifconfig output. > > Sunil Sunder Raj > http://daemon.in > > Hi Sunil, Well the overall usage of mbufs has definitely gone down. But still, there are occasions where the mbufs usage are still above 64K. My ifconfig output shows just my Public static IP and 1 alias. > >From: Tek Bahadur Limbu > >To: freebsd-performance@freebsd.org > >Subject: FreeBSD-6.1 reboots due to high mbufs > >Date: Fri, 30 Mar 2007 12:28:58 +0545 > > > >-----BEGIN PGP SIGNED MESSAGE----- > >Hash: SHA1 > > > >Dear All, > > > >I am quite new to FreeBSD in general and I also don't have much experience > >and knowledge as required by this mailing list. I had posted this on > >freebsd-questions but till now have not received any kind of feedback which > >is fine. > > > >I need some help and suggestions from this mailing list if permissible. I > >am sorry if I am posting this to the wrong mailing list. > > > >I have squid proxy server running on a FreeBSD-6.1 (amd) box. I have been > >facing this problem for sometime now. It's related to mbufs. For some > >reasons, my mbufs usage is extremely high. This high mbufs usage causes > >slow responses from Squid and in rare occasions, it even causes my server > >to reboot or locks out SSH sessions. > > > >This FreeBSD-6.1 squid box serves about 3000-4000 users. > > > >Server specs are: > > > >Dell 430 Intel(R) Pentium(R) D CPU 3.00GHz-Dual CPU > >real memory = 2145959936 (2046 MB) > >bge0: >traffic load: 10Mb/s (through satellite) > > > >root@server# netstat -m > > > >66713/2677/69390 mbufs in use (current/cache/total) > >66686/2210/68896/0 mbuf clusters in use (current/cache/total/max) > >66686/2178 mbuf+clusters out of packet secondary zone in use > >(current/cache) > >0/0/0/0 4k (page size) jumbo clusters in use (current/cache/total/max) > >0/0/0/0 9k jumbo clusters in use (current/cache/total/max) > >0/0/0/0 16k jumbo clusters in use (current/cache/total/max) > >150050K/5089K/155139K bytes allocated to network (current/cache/total) > >0/0/0 requests for mbufs denied (mbufs/clusters/mbuf+clusters) > >0/0/0 requests for jumbo clusters denied (4k/9k/16k) > >0/0/0 sfbufs in use (current/peak/max) > >0 requests for sfbufs denied > >0 requests for sfbufs delayed > >0 requests for I/O initiated by sendfile > >0 calls to protocol drain routines > > > > > >/etc/sysctl.conf: > > > >kern.maxfilesperproc=8192 > >kern.ipc.somaxconn=8192 > >kern.maxprocperuid=8192 > >net.inet.ip.portrange.hifirst=10000 > >net.inet.ip.portrange.first=30000 > >net.inet.ip.portrange.last=65535 > >net.inet.tcp.recvspace=65535 > >net.inet.tcp.sendspace=65535 > >net.inet.tcp.blackhole=2 > >net.inet.udp.blackhole=1 > >net.inet.ip.fw.verbose_limit=5000 > >net.inet.icmp.icmplim=0 > >net.inet.tcp.msl=3000 > >net.isr.direct=1 > >vfs.read_max=16 > > > >/boot/loader.conf: > > > >kern.ipc.nmbclusters=0 > >kern.maxusers=0 > >kern.maxfiles=16384 > >kern.maxproc=8192 > > > >Relevant Kernel Options: > > > >options IPFIREWALL > >options IPFIREWALL_VERBOSE > >options IPFIREWALL_FORWARD > >options IPFIREWALL_VERBOSE_LIMIT=5000 > >options IPFIREWALL_FORWARD_EXTENDED > > > >options MSGMNB=16384 > >options MSGMNI=41 > >options MSGSEG=2049 > >options MSGSSZ=64 > >options MSGTQL=2048 > >options SHMSEG=128 > >options SHMMNI=192 > >options SHMMAX=33554432 > >options SHMMIN=1 > >options SHMALL=8192 > > > >options HZ=1000 #Polling Enabled > > > > > >Note: Running IPFW. Also my other FreeBSD-4.x servers don't seem to suffer > >from this problem. They have almost the same IPFW, Squid and Sysctl > >configs. > > > >Any feedback and suggestions will be highly appreciated. > > > >Thanking you... > > > > > >- -- > > > > > >With best regards and good wishes, > > > >Yours sincerely, > > > >Tek Bahadur Limbu > > > >(TAG/TDG Group) > >Jwl Systems Department > > > >Worldlink Communications Pvt. Ltd. > > > >Jawalakhel, Nepal > > > >http://www.wlink.com.np > >-----BEGIN PGP SIGNATURE----- > >Version: GnuPG v1.4.2.2 (FreeBSD) > > > >iD8DBQFGDLGuVrOl+eVhOvYRAk8dAJ9dumhWCNTp6iHUy5PsQ0ie2OdMqQCgjG8u > >eTxXgRCNSe3R2mmTGSbuSqo= > >=mVGh > >-----END PGP SIGNATURE----- > >_______________________________________________ > >freebsd-performance@freebsd.org mailing list > >http://lists.freebsd.org/mailman/listinfo/freebsd-performance > >To unsubscribe, send any mail to > >"freebsd-performance-unsubscribe@freebsd.org" > > _________________________________________________________________ > Interest Rates Fall Again! $430,000 Mortgage for $1,399/mo - Calculate new > payment > http://www.lowermybills.com/lre/index.jsp?sourceid=lmb-9632-18679&moid=7581 > > - -- With best regards and good wishes, Yours sincerely, Tek Bahadur Limbu (TAG/TDG Group) Jwl Systems Department Worldlink Communications Pvt. Ltd. Jawalakhel, Nepal http://www.wlink.com.np -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2.2 (FreeBSD) iD8DBQFGJJ4uVrOl+eVhOvYRAgAzAKCBdnRkNMTmpbhhIAA3X8z5cz62pACgmFh/ wy7BHBBUth0wp/0+NmMLYbs= =FqgM -----END PGP SIGNATURE----- From owner-freebsd-performance@FreeBSD.ORG Tue Apr 17 16:50:38 2007 Return-Path: X-Original-To: freebsd-performance@freebsd.org Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id BAE8D16A419 for ; Tue, 17 Apr 2007 16:50:38 +0000 (UTC) (envelope-from rwatson@FreeBSD.org) Received: from cyrus.watson.org (cyrus.watson.org [209.31.154.42]) by mx1.freebsd.org (Postfix) with ESMTP id 8271C13C4CB for ; Tue, 17 Apr 2007 16:50:38 +0000 (UTC) (envelope-from rwatson@FreeBSD.org) Received: from fledge.watson.org (fledge.watson.org [209.31.154.41]) by cyrus.watson.org (Postfix) with ESMTP id 0A9AD48415; Tue, 17 Apr 2007 12:50:38 -0400 (EDT) Date: Tue, 17 Apr 2007 17:50:37 +0100 (BST) From: Robert Watson X-X-Sender: robert@fledge.watson.org To: Kevin Way In-Reply-To: <461B0CD0.8090404@insidesystems.net> Message-ID: <20070417174825.X42234@fledge.watson.org> References: <461B0CD0.8090404@insidesystems.net> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Cc: freebsd-performance@freebsd.org Subject: Re: Anyone interested in improving postgresql scaling? 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: Tue, 17 Apr 2007 16:50:38 -0000 On Tue, 10 Apr 2007, Kevin Way wrote: > Kris Kennaway wrote: >> If so, then your task is the following: >> >> Make SYSV semaphores less dumb about process wakeups. Currently whenever >> the semaphore state changes, all processes sleeping on the semaphore are >> woken, even if we only have released enough resources for one waiting >> process to claim. i.e. there is a thundering herd wakeup situation which >> destroys performance at high loads. Fixing this will involve replacing the >> wakeup() calls with appropriate amounts of wakeup_one(). > Could this cause problem cause a situation where an 8-Core system was 50-75% > slower than an otherwise equivalent 2-Core system? > > I have a graph of my sysbench/pgsql results here: > > http://blog.insidesystems.net/files/sysctl-pgsq-amd64-wtf.png > > As the graph shows, the 8-core system is about half the speed of the 2-core > system at 2 simultaneous threads, and it decays down to approximately 1/4 > the speed of the 2-core system as the # of threads hits 5. > > All other (non-pgsql, non-sysv) tests came back approximately as expected, > but I'm left wondering if I did something wrong, or if 8 cpus are slower > than 2, when it comes to Postgres on currently available FreeBSD. > > Kevin Way Inside Systems, Inc. > > (Full detail of what I did is available at: > http://blog.insidesystems.net/articles/2007/04/09/what-did-i-do-wrong ) Would you be able to re-run these tests trying a recent 7.x kernel? If so, make sure you build a non-debugging kernel, and try two variants: one with SCHED_4BSD, and one with SCHED_ULE. Another experiment that would be interesting is to try using device.hints to disable various numbers of CPUs on your 8-core system, and see how the performance graph changes as the number of enabled CPUs changes. Robert N M Watson Computer Laboratory University of Cambridge From owner-freebsd-performance@FreeBSD.ORG Tue Apr 17 18:52:28 2007 Return-Path: X-Original-To: freebsd-performance@freebsd.org Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id D5B8516A400 for ; Tue, 17 Apr 2007 18:52:28 +0000 (UTC) (envelope-from kevin@insidesystems.net) Received: from imap.insidesystems.net (imap.insidesystems.net [206.216.149.56]) by mx1.freebsd.org (Postfix) with ESMTP id AF5FE13C457 for ; Tue, 17 Apr 2007 18:52:28 +0000 (UTC) (envelope-from kevin@insidesystems.net) Received: from [68.32.227.193] (helo=[127.0.0.1]) by imap.insidesystems.net with esmtpa (Exim 4.66 (FreeBSD)) (envelope-from ) id 1Hdsmx-0003kc-N0; Tue, 17 Apr 2007 18:52:27 +0000 Message-ID: <4625174B.5090208@insidesystems.net> Date: Tue, 17 Apr 2007 14:51:55 -0400 From: Kevin Way Organization: Inside Systems, Inc. User-Agent: Thunderbird 2.0.0.0 (Windows/20070326) MIME-Version: 1.0 To: Robert Watson References: <461B0CD0.8090404@insidesystems.net> <20070417174825.X42234@fledge.watson.org> In-Reply-To: <20070417174825.X42234@fledge.watson.org> Content-Type: multipart/signed; protocol="application/x-pkcs7-signature"; micalg=sha1; boundary="------------ms000505020807070508040807" Cc: freebsd-performance@freebsd.org Subject: Re: Anyone interested in improving postgresql scaling? 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: Tue, 17 Apr 2007 18:52:28 -0000 This is a cryptographically signed message in MIME format. --------------ms000505020807070508040807 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Robert Watson wrote: > > On Tue, 10 Apr 2007, Kevin Way wrote: > >> Kris Kennaway wrote: >>> If so, then your task is the following: >>> >>> Make SYSV semaphores less dumb about process wakeups. Currently >>> whenever the semaphore state changes, all processes sleeping on the >>> semaphore are woken, even if we only have released enough resources >>> for one waiting process to claim. i.e. there is a thundering herd >>> wakeup situation which destroys performance at high loads. Fixing >>> this will involve replacing the wakeup() calls with appropriate >>> amounts of wakeup_one(). >> Could this cause problem cause a situation where an 8-Core system was >> 50-75% slower than an otherwise equivalent 2-Core system? >> >> I have a graph of my sysbench/pgsql results here: >> >> http://blog.insidesystems.net/files/sysctl-pgsq-amd64-wtf.png >> >> As the graph shows, the 8-core system is about half the speed of the >> 2-core system at 2 simultaneous threads, and it decays down to >> approximately 1/4 the speed of the 2-core system as the # of threads >> hits 5. >> >> All other (non-pgsql, non-sysv) tests came back approximately as >> expected, but I'm left wondering if I did something wrong, or if 8 >> cpus are slower than 2, when it comes to Postgres on currently >> available FreeBSD. >> >> Kevin Way Inside Systems, Inc. >> >> (Full detail of what I did is available at: >> http://blog.insidesystems.net/articles/2007/04/09/what-did-i-do-wrong ) > > Would you be able to re-run these tests trying a recent 7.x kernel? > If so, make sure you build a non-debugging kernel, and try two > variants: one with SCHED_4BSD, and one with SCHED_ULE. > > Another experiment that would be interesting is to try using > device.hints to disable various numbers of CPUs on your 8-core system, > and see how the performance graph changes as the number of enabled > CPUs changes. > > Robert N M Watson > Computer Laboratory > University of Cambridge I built 7.0 as of 6 days ago, and ran the same test using 8-cores, ULE and 4BSD. The results are available at: http://blog.insidesystems.net/articles/2007/04/11/postgresql-scaling-on-6-2-and-7-0 Unfortunately, I can't run the additional tests, as the machines in question have now been deployed to production. I should have similar equipment available in a few weeks, and will try the other tests at that point. Kevin Way Inside Systems, Inc. --------------ms000505020807070508040807 Content-Type: application/x-pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIMljCC A2IwggLLoAMCAQICEAvaCxfBP4mOqwl0erTOLjMwDQYJKoZIhvcNAQECBQAwXzELMAkGA1UE BhMCVVMxFzAVBgNVBAoTDlZlcmlTaWduLCBJbmMuMTcwNQYDVQQLEy5DbGFzcyAxIFB1Ymxp YyBQcmltYXJ5IENlcnRpZmljYXRpb24gQXV0aG9yaXR5MB4XDTk4MDUxMjAwMDAwMFoXDTA4 MDUxMjIzNTk1OVowgcwxFzAVBgNVBAoTDlZlcmlTaWduLCBJbmMuMR8wHQYDVQQLExZWZXJp U2lnbiBUcnVzdCBOZXR3b3JrMUYwRAYDVQQLEz13d3cudmVyaXNpZ24uY29tL3JlcG9zaXRv cnkvUlBBIEluY29ycC4gQnkgUmVmLixMSUFCLkxURChjKTk4MUgwRgYDVQQDEz9WZXJpU2ln biBDbGFzcyAxIENBIEluZGl2aWR1YWwgU3Vic2NyaWJlci1QZXJzb25hIE5vdCBWYWxpZGF0 ZWQwgZ8wDQYJKoZIhvcNAQEBBQADgY0AMIGJAoGBALtaRIoEFrtV/QN6ii2UTxV4NrgNSrJv nFS/vOh3Kp258Gi7ldkxQXB6gUu5SBNWLccI4YRCq8CikqtEXKpC8IIOAukv+8I7u77JJwpd trA2QjO1blSIT4dKvxna+RXoD4e2HOPMxpqOf2okkuP84GW6p7F+78nbN2rISsgJBuSZAgMB AAGjgbAwga0wDwYDVR0TBAgwBgEB/wIBADBHBgNVHSAEQDA+MDwGC2CGSAGG+EUBBwEBMC0w KwYIKwYBBQUHAgEWH3d3dy52ZXJpc2lnbi5jb20vcmVwb3NpdG9yeS9SUEEwMQYDVR0fBCow KDAmoCSgIoYgaHR0cDovL2NybC52ZXJpc2lnbi5jb20vcGNhMS5jcmwwCwYDVR0PBAQDAgEG MBEGCWCGSAGG+EIBAQQEAwIBBjANBgkqhkiG9w0BAQIFAAOBgQACfZ5vRUs4oLje6VNkIbzk TCuPHv6SQKzYCjlqoTIhLAebq1n+0mIafVU4sDdz3PQHZmNiveFTcFKH56jYUulbLarh3s+s MVTUixnI2COo7wQrMn0sGBzIfImoLnfyRNFlCk10te7TG5JzdC6JOzUTcudAMZrTssSr51a+ i+P7FTCCBJQwggP9oAMCAQICEH//rl4NoatSoVbKWNMj7LQwDQYJKoZIhvcNAQEFBQAwgcwx FzAVBgNVBAoTDlZlcmlTaWduLCBJbmMuMR8wHQYDVQQLExZWZXJpU2lnbiBUcnVzdCBOZXR3 b3JrMUYwRAYDVQQLEz13d3cudmVyaXNpZ24uY29tL3JlcG9zaXRvcnkvUlBBIEluY29ycC4g QnkgUmVmLixMSUFCLkxURChjKTk4MUgwRgYDVQQDEz9WZXJpU2lnbiBDbGFzcyAxIENBIElu ZGl2aWR1YWwgU3Vic2NyaWJlci1QZXJzb25hIE5vdCBWYWxpZGF0ZWQwHhcNMDcwMzA4MDAw MDAwWhcNMDcwNTA3MjM1OTU5WjCCAQYxFzAVBgNVBAoTDlZlcmlTaWduLCBJbmMuMR8wHQYD VQQLExZWZXJpU2lnbiBUcnVzdCBOZXR3b3JrMUYwRAYDVQQLEz13d3cudmVyaXNpZ24uY29t L3JlcG9zaXRvcnkvUlBBIEluY29ycC4gYnkgUmVmLixMSUFCLkxURChjKTk4MR4wHAYDVQQL ExVQZXJzb25hIE5vdCBWYWxpZGF0ZWQxJjAkBgNVBAsTHURpZ2l0YWwgSUQgQ2xhc3MgMSAt IE5ldHNjYXBlMRIwEAYDVQQDFAlLZXZpbiBXYXkxJjAkBgkqhkiG9w0BCQEWF2tldmluQGlu c2lkZXN5c3RlbXMubmV0MIIBIjANBgkqhkiG9w0BAQEFAAOCAQ8AMIIBCgKCAQEArwYwJ9u1 sjZXgnLX1jzAsrIw13m/UOl3OyFkNdnZEo3CH2IiequADpO3FaLLJe7AAIlnP/VDGxv+QHQ1 +eOEu+8jI2xSdTjTMBIxF5fugzEhy1fdQ+QBta85x76+bO4e5XAJUNr8Ao5H8FpCvYlM26hO 5NkUTXdfI7JPszS6AP/e0E6Xrj3zsbZA8HmhOhidH9fG6TNwTPmOG/ji9MvX55OdJcJCIazq cA1kGsgHyX9t62Wn+3NoZBGpFOx+3NVDQglain6VnAHtKwJqcCnRZHtLOO8ofUlZeOR/9beg inkyyKf3F0tyfT8yeprdZ0NreJti9MmhuCj6HmrmO0uGWQIDAQABo4G1MIGyMAkGA1UdEwQC MAAwRAYDVR0gBD0wOzA5BgtghkgBhvhFAQcXAzAqMCgGCCsGAQUFBwIBFhxodHRwczovL3d3 dy52ZXJpc2lnbi5jb20vcnBhMAsGA1UdDwQEAwIFoDAdBgNVHSUEFjAUBggrBgEFBQcDBAYI KwYBBQUHAwIwMwYDVR0fBCwwKjAooCagJIYiaHR0cDovL2NybC52ZXJpc2lnbi5jb20vY2xh c3MxLmNybDANBgkqhkiG9w0BAQUFAAOBgQCJJJAQdU1zobfScAJaaJ76DvJrbJmloOItqhri dCE2396gPTIY9NQjRI3kvbuh1Qzz3voPFeKOlnNXiAJsrwvFCgwFXCxzXZF0J0Wn/Ci6O3YZ BoGIBBgfrt3d5gPkCMTcTDoKqFFLS7fu6eOab8oiGUvKYcXlMxqU8WK4Y3JS2DCCBJQwggP9 oAMCAQICEH//rl4NoatSoVbKWNMj7LQwDQYJKoZIhvcNAQEFBQAwgcwxFzAVBgNVBAoTDlZl cmlTaWduLCBJbmMuMR8wHQYDVQQLExZWZXJpU2lnbiBUcnVzdCBOZXR3b3JrMUYwRAYDVQQL Ez13d3cudmVyaXNpZ24uY29tL3JlcG9zaXRvcnkvUlBBIEluY29ycC4gQnkgUmVmLixMSUFC LkxURChjKTk4MUgwRgYDVQQDEz9WZXJpU2lnbiBDbGFzcyAxIENBIEluZGl2aWR1YWwgU3Vi c2NyaWJlci1QZXJzb25hIE5vdCBWYWxpZGF0ZWQwHhcNMDcwMzA4MDAwMDAwWhcNMDcwNTA3 MjM1OTU5WjCCAQYxFzAVBgNVBAoTDlZlcmlTaWduLCBJbmMuMR8wHQYDVQQLExZWZXJpU2ln biBUcnVzdCBOZXR3b3JrMUYwRAYDVQQLEz13d3cudmVyaXNpZ24uY29tL3JlcG9zaXRvcnkv UlBBIEluY29ycC4gYnkgUmVmLixMSUFCLkxURChjKTk4MR4wHAYDVQQLExVQZXJzb25hIE5v dCBWYWxpZGF0ZWQxJjAkBgNVBAsTHURpZ2l0YWwgSUQgQ2xhc3MgMSAtIE5ldHNjYXBlMRIw EAYDVQQDFAlLZXZpbiBXYXkxJjAkBgkqhkiG9w0BCQEWF2tldmluQGluc2lkZXN5c3RlbXMu bmV0MIIBIjANBgkqhkiG9w0BAQEFAAOCAQ8AMIIBCgKCAQEArwYwJ9u1sjZXgnLX1jzAsrIw 13m/UOl3OyFkNdnZEo3CH2IiequADpO3FaLLJe7AAIlnP/VDGxv+QHQ1+eOEu+8jI2xSdTjT MBIxF5fugzEhy1fdQ+QBta85x76+bO4e5XAJUNr8Ao5H8FpCvYlM26hO5NkUTXdfI7JPszS6 AP/e0E6Xrj3zsbZA8HmhOhidH9fG6TNwTPmOG/ji9MvX55OdJcJCIazqcA1kGsgHyX9t62Wn +3NoZBGpFOx+3NVDQglain6VnAHtKwJqcCnRZHtLOO8ofUlZeOR/9beginkyyKf3F0tyfT8y eprdZ0NreJti9MmhuCj6HmrmO0uGWQIDAQABo4G1MIGyMAkGA1UdEwQCMAAwRAYDVR0gBD0w OzA5BgtghkgBhvhFAQcXAzAqMCgGCCsGAQUFBwIBFhxodHRwczovL3d3dy52ZXJpc2lnbi5j b20vcnBhMAsGA1UdDwQEAwIFoDAdBgNVHSUEFjAUBggrBgEFBQcDBAYIKwYBBQUHAwIwMwYD VR0fBCwwKjAooCagJIYiaHR0cDovL2NybC52ZXJpc2lnbi5jb20vY2xhc3MxLmNybDANBgkq hkiG9w0BAQUFAAOBgQCJJJAQdU1zobfScAJaaJ76DvJrbJmloOItqhridCE2396gPTIY9NQj RI3kvbuh1Qzz3voPFeKOlnNXiAJsrwvFCgwFXCxzXZF0J0Wn/Ci6O3YZBoGIBBgfrt3d5gPk CMTcTDoKqFFLS7fu6eOab8oiGUvKYcXlMxqU8WK4Y3JS2DGCBKowggSmAgEBMIHhMIHMMRcw FQYDVQQKEw5WZXJpU2lnbiwgSW5jLjEfMB0GA1UECxMWVmVyaVNpZ24gVHJ1c3QgTmV0d29y azFGMEQGA1UECxM9d3d3LnZlcmlzaWduLmNvbS9yZXBvc2l0b3J5L1JQQSBJbmNvcnAuIEJ5 IFJlZi4sTElBQi5MVEQoYyk5ODFIMEYGA1UEAxM/VmVyaVNpZ24gQ2xhc3MgMSBDQSBJbmRp dmlkdWFsIFN1YnNjcmliZXItUGVyc29uYSBOb3QgVmFsaWRhdGVkAhB//65eDaGrUqFWyljT I+y0MAkGBSsOAwIaBQCgggKdMBgGCSqGSIb3DQEJAzELBgkqhkiG9w0BBwEwHAYJKoZIhvcN AQkFMQ8XDTA3MDQxNzE4NTE1NVowIwYJKoZIhvcNAQkEMRYEFOoObMuqKk4C9EBlEs6i/K5H hKlEMFIGCSqGSIb3DQEJDzFFMEMwCgYIKoZIhvcNAwcwDgYIKoZIhvcNAwICAgCAMA0GCCqG SIb3DQMCAgFAMAcGBSsOAwIHMA0GCCqGSIb3DQMCAgEoMIHyBgkrBgEEAYI3EAQxgeQwgeEw gcwxFzAVBgNVBAoTDlZlcmlTaWduLCBJbmMuMR8wHQYDVQQLExZWZXJpU2lnbiBUcnVzdCBO ZXR3b3JrMUYwRAYDVQQLEz13d3cudmVyaXNpZ24uY29tL3JlcG9zaXRvcnkvUlBBIEluY29y cC4gQnkgUmVmLixMSUFCLkxURChjKTk4MUgwRgYDVQQDEz9WZXJpU2lnbiBDbGFzcyAxIENB IEluZGl2aWR1YWwgU3Vic2NyaWJlci1QZXJzb25hIE5vdCBWYWxpZGF0ZWQCEH//rl4NoatS oVbKWNMj7LQwgfQGCyqGSIb3DQEJEAILMYHkoIHhMIHMMRcwFQYDVQQKEw5WZXJpU2lnbiwg SW5jLjEfMB0GA1UECxMWVmVyaVNpZ24gVHJ1c3QgTmV0d29yazFGMEQGA1UECxM9d3d3LnZl cmlzaWduLmNvbS9yZXBvc2l0b3J5L1JQQSBJbmNvcnAuIEJ5IFJlZi4sTElBQi5MVEQoYyk5 ODFIMEYGA1UEAxM/VmVyaVNpZ24gQ2xhc3MgMSBDQSBJbmRpdmlkdWFsIFN1YnNjcmliZXIt UGVyc29uYSBOb3QgVmFsaWRhdGVkAhB//65eDaGrUqFWyljTI+y0MA0GCSqGSIb3DQEBAQUA BIIBAHYJxaP0kbCiK9HoRqHy1zeOh7MRxlnZk4+2AXliQw259XvMlgw4ST+4pKZN4s91/beW Y7gdcHevv0fp6/TXX/po2sLm0UoIOp9QKM0u3h8Gdju3Pu/5dJ2TesxwdsCxUlPWszrOHK4E pFpAL7mwviEzLMQ4fNQMS2p/LQh53JI8LCtGYma4hNAfXoFayurZiif/itRT/vOVSb1zyy6B V5XVjDYOeIZMbVkWnlpAzqvUVH2cQlTKv73Uto91JKFQCa8ax7tzMpcVgAeZLozTzGI1MIQO lNh0bcAAqbOQMUFUYnmqUgX7+hmGt19aVkjYpdpJzYYwvMADgsFcVvvEh4UAAAAAAAA= --------------ms000505020807070508040807-- From owner-freebsd-performance@FreeBSD.ORG Tue Apr 17 19:21:21 2007 Return-Path: X-Original-To: freebsd-performance@freebsd.org Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 9171816A555; Tue, 17 Apr 2007 19:21:21 +0000 (UTC) (envelope-from kris@obsecurity.org) Received: from elvis.mu.org (elvis.mu.org [192.203.228.196]) by mx1.freebsd.org (Postfix) with ESMTP id 7CB9913C459; Tue, 17 Apr 2007 19:21:21 +0000 (UTC) (envelope-from kris@obsecurity.org) Received: from obsecurity.dyndns.org (elvis.mu.org [192.203.228.196]) by elvis.mu.org (Postfix) with ESMTP id 1CC3F1A4D83; Tue, 17 Apr 2007 12:21:36 -0700 (PDT) Received: by obsecurity.dyndns.org (Postfix, from userid 1000) id BFA7C51428; Tue, 17 Apr 2007 15:21:20 -0400 (EDT) Date: Tue, 17 Apr 2007 15:21:20 -0400 From: Kris Kennaway To: Kevin Way Message-ID: <20070417192120.GA5191@xor.obsecurity.org> References: <461B0CD0.8090404@insidesystems.net> <20070417174825.X42234@fledge.watson.org> <4625174B.5090208@insidesystems.net> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="G4iJoqBmSsgzjUCe" Content-Disposition: inline In-Reply-To: <4625174B.5090208@insidesystems.net> User-Agent: Mutt/1.4.2.2i Cc: freebsd-performance@freebsd.org, Robert Watson Subject: Re: Anyone interested in improving postgresql scaling? 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: Tue, 17 Apr 2007 19:21:21 -0000 --G4iJoqBmSsgzjUCe Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Tue, Apr 17, 2007 at 02:51:55PM -0400, Kevin Way wrote: > I built 7.0 as of 6 days ago, and ran the same test using 8-cores, ULE=20 > and 4BSD. The results are available at: >=20 > http://blog.insidesystems.net/articles/2007/04/11/postgresql-scaling-on-6= -2-and-7-0 >=20 > Unfortunately, I can't run the additional tests, as the machines in=20 > question have now been deployed to production. I should have similar=20 > equipment available in a few weeks, and will try the other tests at that= =20 > point. That's better but still not comparable to my results. One important change you may have missed was to enable MALLOC_PRODUCTION in /usr/src/lib/libc/stdlib/malloc.c. This disables malloc debugging which may have significant overhead. My results from about a month ago: http://people.freebsd.org/~kris/scaling/nickel.png The big dip between 5-7 clients is believed to be due to a ULE bug that has known cause but is not yet fixed (on dual core CPUs processes are bouncing around between the two cores too much). The main point is that apart from this anomaly in the middle there is approximately linear scaling from 1 to 8 CPUs. On your system it is not scaling linearly beyond 4. I do have some other changes in my kernel but I don't think any of them are enough to explain this difference. Anyway, get back to us when you can retest with the above change. Kris --G4iJoqBmSsgzjUCe Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (FreeBSD) iD4DBQFGJR4wWry0BWjoQKURAq7JAJiPX4ogSZlSl+hTCqAUCLW5zt5TAJ9m28/C ZXwUu7WIgVoXmNrPt8iilA== =c4Xu -----END PGP SIGNATURE----- --G4iJoqBmSsgzjUCe-- From owner-freebsd-performance@FreeBSD.ORG Tue Apr 17 19:34:21 2007 Return-Path: X-Original-To: freebsd-performance@FreeBSD.org Delivered-To: freebsd-performance@FreeBSD.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 52E2D16A407; Tue, 17 Apr 2007 19:34:21 +0000 (UTC) (envelope-from massimo@cedoc.mo.it) Received: from aa014msr.fastwebnet.it (aa014msr.fastwebnet.it [85.18.95.74]) by mx1.freebsd.org (Postfix) with ESMTP id 1453F13C43E; Tue, 17 Apr 2007 19:34:21 +0000 (UTC) (envelope-from massimo@cedoc.mo.it) Received: from intanto (37.254.91.189) by aa014msr.fastwebnet.it (7.3.105.6) id 46249EE2000978DD; Tue, 17 Apr 2007 21:23:07 +0200 Date: Tue, 17 Apr 2007 21:23:07 +0200 From: Massimo Lusetti To: Kevin Way Message-Id: <20070417212307.d01598ee.massimo@cedoc.mo.it> In-Reply-To: <4625174B.5090208@insidesystems.net> References: <461B0CD0.8090404@insidesystems.net> <20070417174825.X42234@fledge.watson.org> <4625174B.5090208@insidesystems.net> X-Mailer: Sylpheed 2.3.1 (GTK+ 2.10.9; i386-portbld-freebsd6.2) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: freebsd-performance@freebsd.org, Robert Watson Subject: Re: Anyone interested in improving postgresql scaling? 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: Tue, 17 Apr 2007 19:34:21 -0000 On Tue, 17 Apr 2007 14:51:55 -0400 Kevin Way wrote: > I built 7.0 as of 6 days ago, and ran the same test using 8-cores, ULE > and 4BSD. The results are available at: > > http://blog.insidesystems.net/articles/2007/04/11/postgresql-scaling-on-6-2-and-7-0 > > Unfortunately, I can't run the additional tests, as the machines in > question have now been deployed to production. I should have similar > equipment available in a few weeks, and will try the other tests at that > point. That's really nice. BTW have you numbers to compare with others OS or DBMS? Regards -- Massimo.run(); I have a hard time being attracted to anyone who can beat me up. -- John McGrath, Atlanta sportswriter, on women weightlifters. From owner-freebsd-performance@FreeBSD.ORG Tue Apr 17 23:17:51 2007 Return-Path: X-Original-To: freebsd-performance@freebsd.org Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 4B0C616A403; Tue, 17 Apr 2007 23:17:51 +0000 (UTC) (envelope-from kevin@insidesystems.net) Received: from imap.insidesystems.net (imap.insidesystems.net [206.216.149.56]) by mx1.freebsd.org (Postfix) with ESMTP id 21A3013C43E; Tue, 17 Apr 2007 23:17:51 +0000 (UTC) (envelope-from kevin@insidesystems.net) Received: from [68.32.227.193] (helo=[127.0.0.1]) by imap.insidesystems.net with esmtpa (Exim 4.66 (FreeBSD)) (envelope-from ) id 1Hdwvk-000Gcl-KQ; Tue, 17 Apr 2007 23:17:48 +0000 Message-ID: <46255581.8000902@insidesystems.net> Date: Tue, 17 Apr 2007 19:17:21 -0400 From: Kevin Way Organization: Inside Systems, Inc. User-Agent: Thunderbird 2.0.0.0 (Windows/20070326) MIME-Version: 1.0 To: Kris Kennaway References: <461B0CD0.8090404@insidesystems.net> <20070417174825.X42234@fledge.watson.org> <4625174B.5090208@insidesystems.net> <20070417192120.GA5191@xor.obsecurity.org> In-Reply-To: <20070417192120.GA5191@xor.obsecurity.org> Content-Type: multipart/signed; protocol="application/x-pkcs7-signature"; micalg=sha1; boundary="------------ms090308060806080007010801" Cc: freebsd-performance@freebsd.org, Robert Watson Subject: Re: Anyone interested in improving postgresql scaling? 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: Tue, 17 Apr 2007 23:17:51 -0000 This is a cryptographically signed message in MIME format. --------------ms090308060806080007010801 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Kris Kennaway wrote: > On Tue, Apr 17, 2007 at 02:51:55PM -0400, Kevin Way wrote: > > >> I built 7.0 as of 6 days ago, and ran the same test using 8-cores, ULE >> and 4BSD. The results are available at: >> >> http://blog.insidesystems.net/articles/2007/04/11/postgresql-scaling-on-6-2-and-7-0 >> >> Unfortunately, I can't run the additional tests, as the machines in >> question have now been deployed to production. I should have similar >> equipment available in a few weeks, and will try the other tests at that >> point. >> > > That's better but still not comparable to my results. One important > change you may have missed was to enable MALLOC_PRODUCTION in > /usr/src/lib/libc/stdlib/malloc.c. This disables malloc debugging > which may have significant overhead. > > My results from about a month ago: > > http://people.freebsd.org/~kris/scaling/nickel.png > > The big dip between 5-7 clients is believed to be due to a ULE bug > that has known cause but is not yet fixed (on dual core CPUs processes > are bouncing around between the two cores too much). The main point > is that apart from this anomaly in the middle there is approximately > linear scaling from 1 to 8 CPUs. On your system it is not scaling > linearly beyond 4. > > I do have some other changes in my kernel but I don't think any of > them are enough to explain this difference. > > Anyway, get back to us when you can retest with the above change. > > Confirmed, I did miss MALLOC_PRODUCTION. I incorrectly believed that I just needed an empty malloc.conf. This is somewhat exciting news from my vantage point. I'll post a corrected re-test as soon as I have another available 8-core machine. --------------ms090308060806080007010801 Content-Type: application/x-pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIMljCC A2IwggLLoAMCAQICEAvaCxfBP4mOqwl0erTOLjMwDQYJKoZIhvcNAQECBQAwXzELMAkGA1UE BhMCVVMxFzAVBgNVBAoTDlZlcmlTaWduLCBJbmMuMTcwNQYDVQQLEy5DbGFzcyAxIFB1Ymxp YyBQcmltYXJ5IENlcnRpZmljYXRpb24gQXV0aG9yaXR5MB4XDTk4MDUxMjAwMDAwMFoXDTA4 MDUxMjIzNTk1OVowgcwxFzAVBgNVBAoTDlZlcmlTaWduLCBJbmMuMR8wHQYDVQQLExZWZXJp U2lnbiBUcnVzdCBOZXR3b3JrMUYwRAYDVQQLEz13d3cudmVyaXNpZ24uY29tL3JlcG9zaXRv cnkvUlBBIEluY29ycC4gQnkgUmVmLixMSUFCLkxURChjKTk4MUgwRgYDVQQDEz9WZXJpU2ln biBDbGFzcyAxIENBIEluZGl2aWR1YWwgU3Vic2NyaWJlci1QZXJzb25hIE5vdCBWYWxpZGF0 ZWQwgZ8wDQYJKoZIhvcNAQEBBQADgY0AMIGJAoGBALtaRIoEFrtV/QN6ii2UTxV4NrgNSrJv nFS/vOh3Kp258Gi7ldkxQXB6gUu5SBNWLccI4YRCq8CikqtEXKpC8IIOAukv+8I7u77JJwpd trA2QjO1blSIT4dKvxna+RXoD4e2HOPMxpqOf2okkuP84GW6p7F+78nbN2rISsgJBuSZAgMB AAGjgbAwga0wDwYDVR0TBAgwBgEB/wIBADBHBgNVHSAEQDA+MDwGC2CGSAGG+EUBBwEBMC0w KwYIKwYBBQUHAgEWH3d3dy52ZXJpc2lnbi5jb20vcmVwb3NpdG9yeS9SUEEwMQYDVR0fBCow KDAmoCSgIoYgaHR0cDovL2NybC52ZXJpc2lnbi5jb20vcGNhMS5jcmwwCwYDVR0PBAQDAgEG MBEGCWCGSAGG+EIBAQQEAwIBBjANBgkqhkiG9w0BAQIFAAOBgQACfZ5vRUs4oLje6VNkIbzk TCuPHv6SQKzYCjlqoTIhLAebq1n+0mIafVU4sDdz3PQHZmNiveFTcFKH56jYUulbLarh3s+s MVTUixnI2COo7wQrMn0sGBzIfImoLnfyRNFlCk10te7TG5JzdC6JOzUTcudAMZrTssSr51a+ i+P7FTCCBJQwggP9oAMCAQICEH//rl4NoatSoVbKWNMj7LQwDQYJKoZIhvcNAQEFBQAwgcwx FzAVBgNVBAoTDlZlcmlTaWduLCBJbmMuMR8wHQYDVQQLExZWZXJpU2lnbiBUcnVzdCBOZXR3 b3JrMUYwRAYDVQQLEz13d3cudmVyaXNpZ24uY29tL3JlcG9zaXRvcnkvUlBBIEluY29ycC4g QnkgUmVmLixMSUFCLkxURChjKTk4MUgwRgYDVQQDEz9WZXJpU2lnbiBDbGFzcyAxIENBIElu ZGl2aWR1YWwgU3Vic2NyaWJlci1QZXJzb25hIE5vdCBWYWxpZGF0ZWQwHhcNMDcwMzA4MDAw MDAwWhcNMDcwNTA3MjM1OTU5WjCCAQYxFzAVBgNVBAoTDlZlcmlTaWduLCBJbmMuMR8wHQYD VQQLExZWZXJpU2lnbiBUcnVzdCBOZXR3b3JrMUYwRAYDVQQLEz13d3cudmVyaXNpZ24uY29t L3JlcG9zaXRvcnkvUlBBIEluY29ycC4gYnkgUmVmLixMSUFCLkxURChjKTk4MR4wHAYDVQQL ExVQZXJzb25hIE5vdCBWYWxpZGF0ZWQxJjAkBgNVBAsTHURpZ2l0YWwgSUQgQ2xhc3MgMSAt IE5ldHNjYXBlMRIwEAYDVQQDFAlLZXZpbiBXYXkxJjAkBgkqhkiG9w0BCQEWF2tldmluQGlu c2lkZXN5c3RlbXMubmV0MIIBIjANBgkqhkiG9w0BAQEFAAOCAQ8AMIIBCgKCAQEArwYwJ9u1 sjZXgnLX1jzAsrIw13m/UOl3OyFkNdnZEo3CH2IiequADpO3FaLLJe7AAIlnP/VDGxv+QHQ1 +eOEu+8jI2xSdTjTMBIxF5fugzEhy1fdQ+QBta85x76+bO4e5XAJUNr8Ao5H8FpCvYlM26hO 5NkUTXdfI7JPszS6AP/e0E6Xrj3zsbZA8HmhOhidH9fG6TNwTPmOG/ji9MvX55OdJcJCIazq cA1kGsgHyX9t62Wn+3NoZBGpFOx+3NVDQglain6VnAHtKwJqcCnRZHtLOO8ofUlZeOR/9beg inkyyKf3F0tyfT8yeprdZ0NreJti9MmhuCj6HmrmO0uGWQIDAQABo4G1MIGyMAkGA1UdEwQC MAAwRAYDVR0gBD0wOzA5BgtghkgBhvhFAQcXAzAqMCgGCCsGAQUFBwIBFhxodHRwczovL3d3 dy52ZXJpc2lnbi5jb20vcnBhMAsGA1UdDwQEAwIFoDAdBgNVHSUEFjAUBggrBgEFBQcDBAYI KwYBBQUHAwIwMwYDVR0fBCwwKjAooCagJIYiaHR0cDovL2NybC52ZXJpc2lnbi5jb20vY2xh c3MxLmNybDANBgkqhkiG9w0BAQUFAAOBgQCJJJAQdU1zobfScAJaaJ76DvJrbJmloOItqhri dCE2396gPTIY9NQjRI3kvbuh1Qzz3voPFeKOlnNXiAJsrwvFCgwFXCxzXZF0J0Wn/Ci6O3YZ BoGIBBgfrt3d5gPkCMTcTDoKqFFLS7fu6eOab8oiGUvKYcXlMxqU8WK4Y3JS2DCCBJQwggP9 oAMCAQICEH//rl4NoatSoVbKWNMj7LQwDQYJKoZIhvcNAQEFBQAwgcwxFzAVBgNVBAoTDlZl cmlTaWduLCBJbmMuMR8wHQYDVQQLExZWZXJpU2lnbiBUcnVzdCBOZXR3b3JrMUYwRAYDVQQL Ez13d3cudmVyaXNpZ24uY29tL3JlcG9zaXRvcnkvUlBBIEluY29ycC4gQnkgUmVmLixMSUFC LkxURChjKTk4MUgwRgYDVQQDEz9WZXJpU2lnbiBDbGFzcyAxIENBIEluZGl2aWR1YWwgU3Vi c2NyaWJlci1QZXJzb25hIE5vdCBWYWxpZGF0ZWQwHhcNMDcwMzA4MDAwMDAwWhcNMDcwNTA3 MjM1OTU5WjCCAQYxFzAVBgNVBAoTDlZlcmlTaWduLCBJbmMuMR8wHQYDVQQLExZWZXJpU2ln biBUcnVzdCBOZXR3b3JrMUYwRAYDVQQLEz13d3cudmVyaXNpZ24uY29tL3JlcG9zaXRvcnkv UlBBIEluY29ycC4gYnkgUmVmLixMSUFCLkxURChjKTk4MR4wHAYDVQQLExVQZXJzb25hIE5v dCBWYWxpZGF0ZWQxJjAkBgNVBAsTHURpZ2l0YWwgSUQgQ2xhc3MgMSAtIE5ldHNjYXBlMRIw EAYDVQQDFAlLZXZpbiBXYXkxJjAkBgkqhkiG9w0BCQEWF2tldmluQGluc2lkZXN5c3RlbXMu bmV0MIIBIjANBgkqhkiG9w0BAQEFAAOCAQ8AMIIBCgKCAQEArwYwJ9u1sjZXgnLX1jzAsrIw 13m/UOl3OyFkNdnZEo3CH2IiequADpO3FaLLJe7AAIlnP/VDGxv+QHQ1+eOEu+8jI2xSdTjT MBIxF5fugzEhy1fdQ+QBta85x76+bO4e5XAJUNr8Ao5H8FpCvYlM26hO5NkUTXdfI7JPszS6 AP/e0E6Xrj3zsbZA8HmhOhidH9fG6TNwTPmOG/ji9MvX55OdJcJCIazqcA1kGsgHyX9t62Wn +3NoZBGpFOx+3NVDQglain6VnAHtKwJqcCnRZHtLOO8ofUlZeOR/9beginkyyKf3F0tyfT8y eprdZ0NreJti9MmhuCj6HmrmO0uGWQIDAQABo4G1MIGyMAkGA1UdEwQCMAAwRAYDVR0gBD0w OzA5BgtghkgBhvhFAQcXAzAqMCgGCCsGAQUFBwIBFhxodHRwczovL3d3dy52ZXJpc2lnbi5j b20vcnBhMAsGA1UdDwQEAwIFoDAdBgNVHSUEFjAUBggrBgEFBQcDBAYIKwYBBQUHAwIwMwYD VR0fBCwwKjAooCagJIYiaHR0cDovL2NybC52ZXJpc2lnbi5jb20vY2xhc3MxLmNybDANBgkq hkiG9w0BAQUFAAOBgQCJJJAQdU1zobfScAJaaJ76DvJrbJmloOItqhridCE2396gPTIY9NQj RI3kvbuh1Qzz3voPFeKOlnNXiAJsrwvFCgwFXCxzXZF0J0Wn/Ci6O3YZBoGIBBgfrt3d5gPk CMTcTDoKqFFLS7fu6eOab8oiGUvKYcXlMxqU8WK4Y3JS2DGCBKowggSmAgEBMIHhMIHMMRcw FQYDVQQKEw5WZXJpU2lnbiwgSW5jLjEfMB0GA1UECxMWVmVyaVNpZ24gVHJ1c3QgTmV0d29y azFGMEQGA1UECxM9d3d3LnZlcmlzaWduLmNvbS9yZXBvc2l0b3J5L1JQQSBJbmNvcnAuIEJ5 IFJlZi4sTElBQi5MVEQoYyk5ODFIMEYGA1UEAxM/VmVyaVNpZ24gQ2xhc3MgMSBDQSBJbmRp dmlkdWFsIFN1YnNjcmliZXItUGVyc29uYSBOb3QgVmFsaWRhdGVkAhB//65eDaGrUqFWyljT I+y0MAkGBSsOAwIaBQCgggKdMBgGCSqGSIb3DQEJAzELBgkqhkiG9w0BBwEwHAYJKoZIhvcN AQkFMQ8XDTA3MDQxNzIzMTcyMVowIwYJKoZIhvcNAQkEMRYEFPhvjMHAXYJwwXzYFx4Ig+5I wtU5MFIGCSqGSIb3DQEJDzFFMEMwCgYIKoZIhvcNAwcwDgYIKoZIhvcNAwICAgCAMA0GCCqG SIb3DQMCAgFAMAcGBSsOAwIHMA0GCCqGSIb3DQMCAgEoMIHyBgkrBgEEAYI3EAQxgeQwgeEw gcwxFzAVBgNVBAoTDlZlcmlTaWduLCBJbmMuMR8wHQYDVQQLExZWZXJpU2lnbiBUcnVzdCBO ZXR3b3JrMUYwRAYDVQQLEz13d3cudmVyaXNpZ24uY29tL3JlcG9zaXRvcnkvUlBBIEluY29y cC4gQnkgUmVmLixMSUFCLkxURChjKTk4MUgwRgYDVQQDEz9WZXJpU2lnbiBDbGFzcyAxIENB IEluZGl2aWR1YWwgU3Vic2NyaWJlci1QZXJzb25hIE5vdCBWYWxpZGF0ZWQCEH//rl4NoatS oVbKWNMj7LQwgfQGCyqGSIb3DQEJEAILMYHkoIHhMIHMMRcwFQYDVQQKEw5WZXJpU2lnbiwg SW5jLjEfMB0GA1UECxMWVmVyaVNpZ24gVHJ1c3QgTmV0d29yazFGMEQGA1UECxM9d3d3LnZl cmlzaWduLmNvbS9yZXBvc2l0b3J5L1JQQSBJbmNvcnAuIEJ5IFJlZi4sTElBQi5MVEQoYyk5 ODFIMEYGA1UEAxM/VmVyaVNpZ24gQ2xhc3MgMSBDQSBJbmRpdmlkdWFsIFN1YnNjcmliZXIt UGVyc29uYSBOb3QgVmFsaWRhdGVkAhB//65eDaGrUqFWyljTI+y0MA0GCSqGSIb3DQEBAQUA BIIBAChHryWsSWWr++Zjz8hDeCM88oyI6XXuSoka3XKSj+78QJOIWg5ecc/jum+ZZIMYs6Q+ Ap0OknKOONVEi5Jb8QimJe+TKtQgqGKejT1YhI7bG4HFo5UWFZJnb/w5pzp1rN7lkOQeL8AQ t/iXQJWcqgbpGeXbeptRWA2SsSiT0LBM59WVTmeNWqz6A9HnxcMsqUwTdEmoAW1MOtxEJJFG YtXqXzhi6vbnf9RKR4+IZ9un1NEELBJvIH2QpyI7quyLV3gpC5cRtVbjXb0W5ZX+LkZseXVI V2Icqj1GeWVuo8DBBz2efZrsl8nj7ci0Y0hvoCKAjhOaPBIRwC7ixMnIJKMAAAAAAAA= --------------ms090308060806080007010801-- From owner-freebsd-performance@FreeBSD.ORG Wed Apr 18 17:58:19 2007 Return-Path: X-Original-To: freebsd-performance@freebsd.org Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 489A116A409 for ; Wed, 18 Apr 2007 17:58:19 +0000 (UTC) (envelope-from knaggsy2000@googlemail.com) Received: from mail10-kcom.uk.cleanport.com (mail10-kcom.uk.cleanport.com [212.79.248.219]) by mx1.freebsd.org (Postfix) with ESMTP id 7E85513C457 for ; Wed, 18 Apr 2007 17:58:18 +0000 (UTC) (envelope-from knaggsy2000@googlemail.com) X-VirusChecked: NOT checked Received: from (unresolved) ([212.50.160.34] HELO=smtpout.karoo.kcom.com) by mail10-kcom.uk.cleanport.com (CleanSMTPd 1.5.5) with ESMTP id 462660B1-0; Wed, 18 Apr 2007 19:58:23 +0200 Received: from [83.100.231.141] (helo=sedna.local) by smtpout.karoo.kcom.comwith esmtp (Exim 4.30) id 1HeEQ3-0007FA-7b server-id smtp-in2; Wed, 18 Apr 2007 18:58:15 +0100 Received: from [10.1.1.1] (asteroid.local [10.1.1.1]) by sedna.local (Postfix) with ESMTP id 4E05C33C35; Wed, 18 Apr 2007 18:03:43 +0100 (BST) Message-ID: <46264F6C.2040803@googlemail.com> Date: Wed, 18 Apr 2007 18:03:40 +0100 From: Danny Knaggs User-Agent: Thunderbird 1.5.0.10 (X11/20070407) MIME-Version: 1.0 To: freebsd-performance@freebsd.org References: <46160016.8080504@googlemail.com> <70e8236f0704090410r5457e32ct3928edea654868b6@mail.gmail.com> In-Reply-To: <70e8236f0704090410r5457e32ct3928edea654868b6@mail.gmail.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: quoted-printable Cc: joao.barros@gmail.com, sos@freebsd.org Subject: Re: FreeBSD 6.2 on SPARC64/x86 with Promise IDE Controller 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, 18 Apr 2007 17:58:19 -0000 Any news/updates on this? Every thing has gone silent! Joao Barros wrote: > I think nullpt sent you off track when he told you to try=20 > freebsd-performance. > I believe this is an issue for drivers and/or sparc (or maybe current) >=20 > I think I have a TX2 too. I can try it on my Ultra 5 if more testing > is required. >=20 > PS: I'm cc'ing S=F8ren Schmidt, he's the maintainer of most of the ata=20 > drivers >=20 >=20 > On 4/6/07, Danny Knaggs wrote: >> Hello all! >> >> This is first time I've used a mailing list, so bear with me! >> >> >> I've been asked to submit my findings of the ata driver in FreeBSD 6.2 >> on my sparc64 and x86 box from bsdforums.org. >> >> Link to my thread: http://www.bsdforums.org/forums/showthread.php?t=3D= 48682 >> >> >> I've just installed a Promise IDE Controller card (Ultra 133 TX2 - >> PDC20269) in my Sun Ultra 10 and have come across a slight snag. >> >> If I don't put in "hw.ata.ata_dma=3D0" in the loader options I get DMA >> timeout errors after it has queried the HDD on the Promise controller. >> >> I have found a link which someone else has a similar problem (NetBSD o= n >> Alpha) which maybe useful: >> http://archive.netbsd.se/?ml=3Dfreebsd-alpha&a=3D2007-02&t=3D3177803 >> >> Now, after BSD has loaded I can successfully change the DMA mode to >> UDMA66 on the HDD without any problems (get ~30MB/s transfer rate, >> compared to ~15MB/s when using the on-board controller using "dd"). An= y >> higher and I get DMA timeout messages. >> >> The HDD works fine when it's attached to the on-board controller. >> >> >> Now, I thought I try the same Promise IDE card in x86 box with FreeBSD >> 6.2 and found something interesting... >> >> The HDD will not operate correctly at UDMA133 - it performs very slowl= y >> (<15Mb/s). >> >> Forcing the HDD to run at UDMA100 gives me 64Mb/s transfer. Which is >> roughly what I expect. >> >> >> So, it seems something is broken with the ATA driver - Sparc/Alpha >> getting the worse of it! >> >> If anyone has ideas/brainwaves/etc - I'm willing to give it a whirl! >> >> >> Thanks in advance. >> >> >> Dan. >> >> _______________________________________________ >> freebsd-performance@freebsd.org mailing list >> http://lists.freebsd.org/mailman/listinfo/freebsd-performance >> To unsubscribe, send any mail to=20 >> "freebsd-performance-unsubscribe@freebsd.org" >> >=20 >=20 From owner-freebsd-performance@FreeBSD.ORG Thu Apr 19 03:05:47 2007 Return-Path: X-Original-To: freebsd-performance@freebsd.org Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id A93E116A401 for ; Thu, 19 Apr 2007 03:05:47 +0000 (UTC) (envelope-from mv@thebeastie.org) Received: from p4.roq.com (ns1.ecoms.com [207.44.130.137]) by mx1.freebsd.org (Postfix) with ESMTP id 839A713C455 for ; Thu, 19 Apr 2007 03:05:47 +0000 (UTC) (envelope-from mv@thebeastie.org) Received: from p4.roq.com (localhost.roq.com [127.0.0.1]) by p4.roq.com (Postfix) with ESMTP id F281C4CB85; Thu, 19 Apr 2007 02:33:19 +0000 (GMT) Received: from smitch7.jumbuck.com (p82.jumbuck.com [206.112.99.82]) by p4.roq.com (Postfix) with ESMTP id C041A4CB44; Thu, 19 Apr 2007 02:33:19 +0000 (GMT) Received: from smitch7.jumbuck.com (mail.jumbuck.com [206.112.99.82]) by smitch7.jumbuck.com (Postfix) with ESMTP id 3BBBC411299; Thu, 19 Apr 2007 02:32:56 +0000 (UTC) Received: from beaste5.jumbuck.com (melbourne.jumbuck.com [150.101.166.27]) (using TLSv1 with cipher ADH-AES256-SHA (256/256 bits)) (No client certificate requested) by smitch7.jumbuck.com (Postfix) with ESMTP id DF17B411248; Thu, 19 Apr 2007 02:32:55 +0000 (UTC) Received: from beaste5.jumbuck.com (beast5 [192.168.46.105]) by beaste5.jumbuck.com (Postfix) with ESMTP id BD310209D1A9; Thu, 19 Apr 2007 12:32:54 +1000 (EST) Received: from beast2.jumbuck.com (unknown [192.168.46.102]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by beaste5.jumbuck.com (Postfix) with ESMTP id 8E2E8209D195; Thu, 19 Apr 2007 12:32:54 +1000 (EST) Message-ID: <4626D4D5.7030303@thebeastie.org> Date: Thu, 19 Apr 2007 12:32:53 +1000 From: Michael Vince User-Agent: Mozilla/5.0 (X11; U; FreeBSD amd64; en-US; rv:1.8.1.2) Gecko/20070314 SeaMonkey/1.1.1 MIME-Version: 1.0 To: freebsd-java@FreeBSD.org Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Virus-Scanned: ClamAV using ClamSMTP Cc: freebsd-performance@freebsd.org Subject: Java performance on AMD64 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: Thu, 19 Apr 2007 03:05:47 -0000 Hey All, I been benchmarking Diablo Java under AMD64 on 6.2R and using the same methods I posted a while ago detailed somewhat here http://lists.freebsd.org/pipermail/freebsd-java/2006-August/005576.html The difference here is that libthr now works under Amd64. But it appears libthr to be about half the speed of libpthread under AMD64 mode which is the opposite behavior when using it under i386. The only other difference is that I am now using a 4 core server (total) Intel Core2 cpus, I tried testing under a dual core (total) server and the performance was just about exactly neck and neck between libpthread and libthr which is also somewhat wierd. The servers are under stock 6.2R and I did have a multi IP jail patch applied and I tried removing that and it had no difference on performance. Does any one know with there is such a large amount of difference of behavior between i386 and Amd64 on Java? Thanks, Mike From owner-freebsd-performance@FreeBSD.ORG Thu Apr 19 03:19:14 2007 Return-Path: X-Original-To: freebsd-performance@freebsd.org Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id CF35C16A404; Thu, 19 Apr 2007 03:19:14 +0000 (UTC) (envelope-from kris@obsecurity.org) Received: from elvis.mu.org (elvis.mu.org [192.203.228.196]) by mx1.freebsd.org (Postfix) with ESMTP id BECA813C46A; Thu, 19 Apr 2007 03:19:14 +0000 (UTC) (envelope-from kris@obsecurity.org) Received: from obsecurity.dyndns.org (elvis.mu.org [192.203.228.196]) by elvis.mu.org (Postfix) with ESMTP id 105E31A4D83; Wed, 18 Apr 2007 20:19:31 -0700 (PDT) Received: by obsecurity.dyndns.org (Postfix, from userid 1000) id EB65851410; Wed, 18 Apr 2007 23:19:13 -0400 (EDT) Date: Wed, 18 Apr 2007 23:19:13 -0400 From: Kris Kennaway To: Michael Vince Message-ID: <20070419031913.GA48411@xor.obsecurity.org> References: <4626D4D5.7030303@thebeastie.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4626D4D5.7030303@thebeastie.org> User-Agent: Mutt/1.4.2.2i Cc: freebsd-performance@freebsd.org, freebsd-java@FreeBSD.org Subject: Re: Java performance on AMD64 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: Thu, 19 Apr 2007 03:19:14 -0000 On Thu, Apr 19, 2007 at 12:32:53PM +1000, Michael Vince wrote: > Hey All, > I been benchmarking Diablo Java under AMD64 on 6.2R and using the same > methods I posted a while ago detailed somewhat here > http://lists.freebsd.org/pipermail/freebsd-java/2006-August/005576.html > > The difference here is that libthr now works under Amd64. > But it appears libthr to be about half the speed of libpthread under > AMD64 mode which is the opposite behavior when using it under i386. That's contrary to my benchmarks :( Dunno what might be wrong though, assuming you have checked all the obvious. Kris From owner-freebsd-performance@FreeBSD.ORG Fri Apr 20 06:59:17 2007 Return-Path: X-Original-To: freebsd-performance@freebsd.org Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 4857316A401; Fri, 20 Apr 2007 06:59:17 +0000 (UTC) (envelope-from mv@thebeastie.org) Received: from p4.roq.com (ns1.ecoms.com [207.44.130.137]) by mx1.freebsd.org (Postfix) with ESMTP id 20A9013C468; Fri, 20 Apr 2007 06:59:17 +0000 (UTC) (envelope-from mv@thebeastie.org) Received: from p4.roq.com (localhost.roq.com [127.0.0.1]) by p4.roq.com (Postfix) with ESMTP id 5263C4CE26; Fri, 20 Apr 2007 06:59:41 +0000 (GMT) Received: from smitch7.jumbuck.com (p82.jumbuck.com [206.112.99.82]) by p4.roq.com (Postfix) with ESMTP id 295404CE00; Fri, 20 Apr 2007 06:59:41 +0000 (GMT) Received: from smitch7.jumbuck.com (mail.jumbuck.com [206.112.99.82]) by smitch7.jumbuck.com (Postfix) with ESMTP id E67D2410EB0; Fri, 20 Apr 2007 06:59:15 +0000 (UTC) Received: from beaste5.jumbuck.com (melbourne.jumbuck.com [150.101.166.27]) (using TLSv1 with cipher ADH-AES256-SHA (256/256 bits)) (No client certificate requested) by smitch7.jumbuck.com (Postfix) with ESMTP id 97508410D44; Fri, 20 Apr 2007 06:59:15 +0000 (UTC) Received: from beaste5.jumbuck.com (beast5 [192.168.46.105]) by beaste5.jumbuck.com (Postfix) with ESMTP id 24C05209D265; Fri, 20 Apr 2007 16:59:13 +1000 (EST) Received: from beast2.jumbuck.com (unknown [192.168.46.102]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by beaste5.jumbuck.com (Postfix) with ESMTP id F0F4B209D1CE; Fri, 20 Apr 2007 16:59:12 +1000 (EST) Message-ID: <462864C0.7060100@thebeastie.org> Date: Fri, 20 Apr 2007 16:59:12 +1000 From: Michael Vince User-Agent: Mozilla/5.0 (X11; U; FreeBSD amd64; en-US; rv:1.8.1.2) Gecko/20070314 SeaMonkey/1.1.1 MIME-Version: 1.0 To: Kris Kennaway References: <4626D4D5.7030303@thebeastie.org> <20070419031913.GA48411@xor.obsecurity.org> In-Reply-To: <20070419031913.GA48411@xor.obsecurity.org> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Virus-Scanned: ClamAV using ClamSMTP Cc: freebsd-performance@freebsd.org, freebsd-java@FreeBSD.org Subject: Re: Java performance on AMD64 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: Fri, 20 Apr 2007 06:59:17 -0000 Kris Kennaway wrote: > On Thu, Apr 19, 2007 at 12:32:53PM +1000, Michael Vince wrote: > >> Hey All, >> I been benchmarking Diablo Java under AMD64 on 6.2R and using the same >> methods I posted a while ago detailed somewhat here >> http://lists.freebsd.org/pipermail/freebsd-java/2006-August/005576.html >> >> The difference here is that libthr now works under Amd64. >> But it appears libthr to be about half the speed of libpthread under >> AMD64 mode which is the opposite behavior when using it under i386. >> > > That's contrary to my benchmarks :( Dunno what might be wrong though, > assuming you have checked all the obvious. > > Kris > OK, I did more testing and it appears the 6.1R Diablo Java binary package on Amd64 on 6.2/libthr appears to be the problem, it's as much as 60% slower then libpthread. On 6.2R Amd64 I built the ports version of jdk1.5 and libthr appeared to be around 25% faster then libpthread. But using the Diablo jdk package for 6.1R on 6.2 under libpthread is even faster by around 15% over anything I could do for any combination of Java and libthr. I would guess that a new 6.2R Amd64 Diablo package would probably turn it around again for libthr and I would guess it would again around 20% faster then anything I can get from Java under AMD64 6.2R, what's going on here exactly I don't know. Cheers, Mike From owner-freebsd-performance@FreeBSD.ORG Fri Apr 20 07:09:35 2007 Return-Path: X-Original-To: freebsd-performance@freebsd.org Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 2764016A414; Fri, 20 Apr 2007 07:09:35 +0000 (UTC) (envelope-from kris@obsecurity.org) Received: from elvis.mu.org (elvis.mu.org [192.203.228.196]) by mx1.freebsd.org (Postfix) with ESMTP id 138A613C48C; Fri, 20 Apr 2007 07:09:35 +0000 (UTC) (envelope-from kris@obsecurity.org) Received: from obsecurity.dyndns.org (elvis.mu.org [192.203.228.196]) by elvis.mu.org (Postfix) with ESMTP id DA5C31A4D91; Fri, 20 Apr 2007 00:09:52 -0700 (PDT) Received: by obsecurity.dyndns.org (Postfix, from userid 1000) id 33E9B5138E; Fri, 20 Apr 2007 03:09:34 -0400 (EDT) Date: Fri, 20 Apr 2007 03:09:34 -0400 From: Kris Kennaway To: Michael Vince Message-ID: <20070420070934.GA5367@xor.obsecurity.org> References: <4626D4D5.7030303@thebeastie.org> <20070419031913.GA48411@xor.obsecurity.org> <462864C0.7060100@thebeastie.org> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="tKW2IUtsqtDRztdT" Content-Disposition: inline In-Reply-To: <462864C0.7060100@thebeastie.org> User-Agent: Mutt/1.4.2.2i Cc: freebsd-performance@freebsd.org, freebsd-java@FreeBSD.org, Kris Kennaway Subject: Re: Java performance on AMD64 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: Fri, 20 Apr 2007 07:09:35 -0000 --tKW2IUtsqtDRztdT Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Fri, Apr 20, 2007 at 04:59:12PM +1000, Michael Vince wrote: > Kris Kennaway wrote: > >On Thu, Apr 19, 2007 at 12:32:53PM +1000, Michael Vince wrote: > > =20 > >>Hey All, > >>I been benchmarking Diablo Java under AMD64 on 6.2R and using the same= =20 > >>methods I posted a while ago detailed somewhat here > >>http://lists.freebsd.org/pipermail/freebsd-java/2006-August/005576.html > >> > >>The difference here is that libthr now works under Amd64. > >>But it appears libthr to be about half the speed of libpthread under=20 > >>AMD64 mode which is the opposite behavior when using it under i386. > >> =20 > > > >That's contrary to my benchmarks :( Dunno what might be wrong though, > >assuming you have checked all the obvious. > > > >Kris > > =20 > OK, > I did more testing and it appears the 6.1R Diablo Java binary package on= =20 > Amd64 on 6.2/libthr appears to be the problem, it's as much as 60%=20 > slower then libpthread. >=20 > On 6.2R Amd64 I built the ports version of jdk1.5 and libthr appeared to= =20 > be around 25% faster then libpthread. > But using the Diablo jdk package for 6.1R on 6.2 under libpthread is=20 > even faster by around 15% over anything I could do for any combination=20 > of Java and libthr. >=20 > I would guess that a new 6.2R Amd64 Diablo package would probably turn=20 > it around again for libthr and I would guess it would again around 20%=20 > faster then anything I can get from Java under AMD64 6.2R, what's going= =20 > on here exactly I don't know. That's weird, I thought the diablo package was just compiled from jdk1.5. It could be there was a performance regression from a change made to the port after 6.1 - it would be great if you can follow it up with the java@ people. Kris --tKW2IUtsqtDRztdT Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (FreeBSD) iD8DBQFGKGctWry0BWjoQKURAjlxAKCd7A2lfo107ZeofW5Z/y9Vr8kCyACggc0M 3ihbYuAXR2EssgzZN8o7378= =Stgd -----END PGP SIGNATURE----- --tKW2IUtsqtDRztdT-- From owner-freebsd-performance@FreeBSD.ORG Fri Apr 20 19:22:28 2007 Return-Path: X-Original-To: freebsd-performance@freebsd.org Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 9883216A401 for ; Fri, 20 Apr 2007 19:22:28 +0000 (UTC) (envelope-from glewis@eyesbeyond.com) Received: from misty.eyesbeyond.com (gerbercreations.com [71.39.140.16]) by mx1.freebsd.org (Postfix) with ESMTP id 140CC13C487 for ; Fri, 20 Apr 2007 19:22:27 +0000 (UTC) (envelope-from glewis@eyesbeyond.com) Received: from misty.eyesbeyond.com (localhost.eyesbeyond.com [127.0.0.1]) by misty.eyesbeyond.com (8.13.1/8.13.3) with ESMTP id l3KIpSq4024042; Fri, 20 Apr 2007 11:51:28 -0700 (PDT) (envelope-from glewis@eyesbeyond.com) Received: (from glewis@localhost) by misty.eyesbeyond.com (8.13.1/8.13.3/Submit) id l3KIpP1Z024041; Fri, 20 Apr 2007 11:51:25 -0700 (PDT) (envelope-from glewis@eyesbeyond.com) X-Authentication-Warning: misty.eyesbeyond.com: glewis set sender to glewis@eyesbeyond.com using -f Date: Fri, 20 Apr 2007 11:51:24 -0700 From: Greg Lewis To: Kris Kennaway Message-ID: <20070420185124.GA23914@misty.eyesbeyond.com> References: <4626D4D5.7030303@thebeastie.org> <20070419031913.GA48411@xor.obsecurity.org> <462864C0.7060100@thebeastie.org> <20070420070934.GA5367@xor.obsecurity.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20070420070934.GA5367@xor.obsecurity.org> User-Agent: Mutt/1.4.2.2i X-Mailman-Approved-At: Fri, 20 Apr 2007 19:48:51 +0000 Cc: freebsd-performance@freebsd.org, Michael Vince , freebsd-java@freebsd.org Subject: Re: Java performance on AMD64 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: Fri, 20 Apr 2007 19:22:28 -0000 On Fri, Apr 20, 2007 at 03:09:34AM -0400, Kris Kennaway wrote: > On Fri, Apr 20, 2007 at 04:59:12PM +1000, Michael Vince wrote: > > Kris Kennaway wrote: > > >On Thu, Apr 19, 2007 at 12:32:53PM +1000, Michael Vince wrote: > > > > > >>Hey All, > > >>I been benchmarking Diablo Java under AMD64 on 6.2R and using the same > > >>methods I posted a while ago detailed somewhat here > > >>http://lists.freebsd.org/pipermail/freebsd-java/2006-August/005576.html > > >> > > >>The difference here is that libthr now works under Amd64. > > >>But it appears libthr to be about half the speed of libpthread under > > >>AMD64 mode which is the opposite behavior when using it under i386. > > >> > > > > > >That's contrary to my benchmarks :( Dunno what might be wrong though, > > >assuming you have checked all the obvious. > > > > > >Kris > > > > > OK, > > I did more testing and it appears the 6.1R Diablo Java binary package on > > Amd64 on 6.2/libthr appears to be the problem, it's as much as 60% > > slower then libpthread. > > > > On 6.2R Amd64 I built the ports version of jdk1.5 and libthr appeared to > > be around 25% faster then libpthread. > > But using the Diablo jdk package for 6.1R on 6.2 under libpthread is > > even faster by around 15% over anything I could do for any combination > > of Java and libthr. > > > > I would guess that a new 6.2R Amd64 Diablo package would probably turn > > it around again for libthr and I would guess it would again around 20% > > faster then anything I can get from Java under AMD64 6.2R, what's going > > on here exactly I don't know. > > That's weird, I thought the diablo package was just compiled from > jdk1.5. It could be there was a performance regression from a change > made to the port after 6.1 - it would be great if you can follow it up > with the java@ people. Diablo is compiled from the partner source. jdk15 is built on the SCSL release. The SCSL release was done when 1.5.0 was released. The current release of Diablo is based off of 1.5.0_07, so there were changes in between the two that may/will affect performance. -- Greg Lewis Email : glewis@eyesbeyond.com Eyes Beyond Web : http://www.eyesbeyond.com Information Technology FreeBSD : glewis@FreeBSD.org From owner-freebsd-performance@FreeBSD.ORG Fri Apr 20 19:29:20 2007 Return-Path: X-Original-To: freebsd-performance@freebsd.org Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id AC45216A401; Fri, 20 Apr 2007 19:29:20 +0000 (UTC) (envelope-from sean@mcneil.com) Received: from mail.mcneil.com (mcneil.com [24.199.45.54]) by mx1.freebsd.org (Postfix) with ESMTP id 6AC3413C501; Fri, 20 Apr 2007 19:29:20 +0000 (UTC) (envelope-from sean@mcneil.com) Received: from localhost (localhost.mcneil.com [127.0.0.1]) by mail.mcneil.com (Postfix) with ESMTP id 417DDF28A9; Fri, 20 Apr 2007 12:13:44 -0700 (PDT) X-Virus-Scanned: amavisd-new at mcneil.com Received: from mail.mcneil.com ([127.0.0.1]) by localhost (mcneil.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id hN4tUtSfrNVZ; Fri, 20 Apr 2007 12:13:42 -0700 (PDT) Received: from [10.1.0.60] (mini.mcneil.com [10.1.0.60]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by mail.mcneil.com (Postfix) with ESMTP id 8A97BF28A4; Fri, 20 Apr 2007 12:13:42 -0700 (PDT) Message-ID: <462910CF.8050602@mcneil.com> Date: Fri, 20 Apr 2007 12:13:19 -0700 From: Sean McNeil Organization: Sean McNeil Consulting, Inc User-Agent: Thunderbird 2.0.0.0 (Windows/20070326) MIME-Version: 1.0 To: Greg Lewis References: <4626D4D5.7030303@thebeastie.org> <20070419031913.GA48411@xor.obsecurity.org> <462864C0.7060100@thebeastie.org> <20070420070934.GA5367@xor.obsecurity.org> <20070420185124.GA23914@misty.eyesbeyond.com> In-Reply-To: <20070420185124.GA23914@misty.eyesbeyond.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Mailman-Approved-At: Fri, 20 Apr 2007 19:49:08 +0000 Cc: freebsd-performance@freebsd.org, freebsd-java@freebsd.org, Michael Vince , Kris Kennaway Subject: Re: Java performance on AMD64 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: Fri, 20 Apr 2007 19:29:20 -0000 Greg Lewis wrote: > On Fri, Apr 20, 2007 at 03:09:34AM -0400, Kris Kennaway wrote: > >> On Fri, Apr 20, 2007 at 04:59:12PM +1000, Michael Vince wrote: >> >>> Kris Kennaway wrote: >>> >>>> On Thu, Apr 19, 2007 at 12:32:53PM +1000, Michael Vince wrote: >>>> >>>> >>>>> Hey All, >>>>> I been benchmarking Diablo Java under AMD64 on 6.2R and using the same >>>>> methods I posted a while ago detailed somewhat here >>>>> http://lists.freebsd.org/pipermail/freebsd-java/2006-August/005576.html >>>>> >>>>> The difference here is that libthr now works under Amd64. >>>>> But it appears libthr to be about half the speed of libpthread under >>>>> AMD64 mode which is the opposite behavior when using it under i386. >>>>> >>>>> >>>> That's contrary to my benchmarks :( Dunno what might be wrong though, >>>> assuming you have checked all the obvious. >>>> >>>> Kris >>>> >>>> >>> OK, >>> I did more testing and it appears the 6.1R Diablo Java binary package on >>> Amd64 on 6.2/libthr appears to be the problem, it's as much as 60% >>> slower then libpthread. >>> >>> On 6.2R Amd64 I built the ports version of jdk1.5 and libthr appeared to >>> be around 25% faster then libpthread. >>> But using the Diablo jdk package for 6.1R on 6.2 under libpthread is >>> even faster by around 15% over anything I could do for any combination >>> of Java and libthr. >>> >>> I would guess that a new 6.2R Amd64 Diablo package would probably turn >>> it around again for libthr and I would guess it would again around 20% >>> faster then anything I can get from Java under AMD64 6.2R, what's going >>> on here exactly I don't know. >>> >> That's weird, I thought the diablo package was just compiled from >> jdk1.5. It could be there was a performance regression from a change >> made to the port after 6.1 - it would be great if you can follow it up >> with the java@ people. >> > > Diablo is compiled from the partner source. jdk15 is built on the SCSL > release. The SCSL release was done when 1.5.0 was released. The current > release of Diablo is based off of 1.5.0_07, so there were changes in between > the two that may/will affect performance. > I am confused why this is thought to be a problem with Java. Is it not true that libthr does not allow new threads to be scheduled on other CPUs? I thought that was the whole point of the difference between libthr and libpthread. This would mean that having additional CPUs would be beneficial to libpthread whereas it would not help libthr performance. Is all your testing being done on machines with the same number of CPUs? From owner-freebsd-performance@FreeBSD.ORG Fri Apr 20 20:32:02 2007 Return-Path: X-Original-To: freebsd-performance@freebsd.org Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 53E1616A407; Fri, 20 Apr 2007 20:32:02 +0000 (UTC) (envelope-from kris@obsecurity.org) Received: from elvis.mu.org (elvis.mu.org [192.203.228.196]) by mx1.freebsd.org (Postfix) with ESMTP id 3F66713C489; Fri, 20 Apr 2007 20:32:02 +0000 (UTC) (envelope-from kris@obsecurity.org) Received: from obsecurity.dyndns.org (elvis.mu.org [192.203.228.196]) by elvis.mu.org (Postfix) with ESMTP id 8B04A1A4D9D; Fri, 20 Apr 2007 13:32:20 -0700 (PDT) Received: by obsecurity.dyndns.org (Postfix, from userid 1000) id 84B9D51387; Fri, 20 Apr 2007 16:32:00 -0400 (EDT) Date: Fri, 20 Apr 2007 16:32:00 -0400 From: Kris Kennaway To: Sean McNeil Message-ID: <20070420203200.GA39406@xor.obsecurity.org> References: <4626D4D5.7030303@thebeastie.org> <20070419031913.GA48411@xor.obsecurity.org> <462864C0.7060100@thebeastie.org> <20070420070934.GA5367@xor.obsecurity.org> <20070420185124.GA23914@misty.eyesbeyond.com> <462910CF.8050602@mcneil.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <462910CF.8050602@mcneil.com> User-Agent: Mutt/1.4.2.2i Cc: Greg Lewis , freebsd-performance@freebsd.org, freebsd-java@freebsd.org, Michael Vince , Kris Kennaway Subject: Re: Java performance on AMD64 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: Fri, 20 Apr 2007 20:32:02 -0000 On Fri, Apr 20, 2007 at 12:13:19PM -0700, Sean McNeil wrote: > Greg Lewis wrote: > >On Fri, Apr 20, 2007 at 03:09:34AM -0400, Kris Kennaway wrote: > > > >>On Fri, Apr 20, 2007 at 04:59:12PM +1000, Michael Vince wrote: > >> > >>>Kris Kennaway wrote: > >>> > >>>>On Thu, Apr 19, 2007 at 12:32:53PM +1000, Michael Vince wrote: > >>>> > >>>> > >>>>>Hey All, > >>>>>I been benchmarking Diablo Java under AMD64 on 6.2R and using the same > >>>>>methods I posted a while ago detailed somewhat here > >>>>>http://lists.freebsd.org/pipermail/freebsd-java/2006-August/005576.html > >>>>> > >>>>>The difference here is that libthr now works under Amd64. > >>>>>But it appears libthr to be about half the speed of libpthread under > >>>>>AMD64 mode which is the opposite behavior when using it under i386. > >>>>> > >>>>> > >>>>That's contrary to my benchmarks :( Dunno what might be wrong though, > >>>>assuming you have checked all the obvious. > >>>> > >>>>Kris > >>>> > >>>> > >>>OK, > >>>I did more testing and it appears the 6.1R Diablo Java binary package on > >>>Amd64 on 6.2/libthr appears to be the problem, it's as much as 60% > >>>slower then libpthread. > >>> > >>>On 6.2R Amd64 I built the ports version of jdk1.5 and libthr appeared to > >>>be around 25% faster then libpthread. > >>>But using the Diablo jdk package for 6.1R on 6.2 under libpthread is > >>>even faster by around 15% over anything I could do for any combination > >>>of Java and libthr. > >>> > >>>I would guess that a new 6.2R Amd64 Diablo package would probably turn > >>>it around again for libthr and I would guess it would again around 20% > >>>faster then anything I can get from Java under AMD64 6.2R, what's going > >>>on here exactly I don't know. > >>> > >>That's weird, I thought the diablo package was just compiled from > >>jdk1.5. It could be there was a performance regression from a change > >>made to the port after 6.1 - it would be great if you can follow it up > >>with the java@ people. > >> > > > >Diablo is compiled from the partner source. jdk15 is built on the SCSL > >release. The SCSL release was done when 1.5.0 was released. The current > >release of Diablo is based off of 1.5.0_07, so there were changes in > >between > >the two that may/will affect performance. > > > I am confused why this is thought to be a problem with Java. Is it not > true that libthr does not allow new threads to be scheduled on other > CPUs? Yes, that is not true. You are confusing it with the old libc_r. Kris From owner-freebsd-performance@FreeBSD.ORG Sat Apr 21 03:12:35 2007 Return-Path: X-Original-To: freebsd-performance@FreeBSD.org Delivered-To: freebsd-performance@FreeBSD.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 9A73416A404 for ; Sat, 21 Apr 2007 03:12:35 +0000 (UTC) (envelope-from alepulver@FreeBSD.org) Received: from relay03.pair.com (relay03.pair.com [209.68.5.17]) by mx1.freebsd.org (Postfix) with SMTP id 52BB513C483 for ; Sat, 21 Apr 2007 03:12:35 +0000 (UTC) (envelope-from alepulver@FreeBSD.org) Received: (qmail 50561 invoked by uid 0); 21 Apr 2007 03:05:53 -0000 Received: from 190.55.91.88 (HELO deimos.mars.bsd) (190.55.91.88) by relay03.pair.com with SMTP; 21 Apr 2007 03:05:53 -0000 X-pair-Authenticated: 190.55.91.88 Date: Sat, 21 Apr 2007 00:05:29 -0300 From: Alejandro Pulver To: freebsd-performance@FreeBSD.org Message-ID: <20070421000529.1ccdf20b@deimos.mars.bsd> X-Mailer: Claws Mail 2.8.1 (GTK+ 2.10.11; i386-portbld-freebsd6.2) Mime-Version: 1.0 Content-Type: multipart/signed; boundary=Sig_Jip+y.ksvo1mSrs3gahAUT8; protocol="application/pgp-signature"; micalg=PGP-SHA1 X-Mailman-Approved-At: Sat, 21 Apr 2007 04:41:32 +0000 Cc: Subject: sysutils/fusefs-ntfs: slow reading/writing speed 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: Sat, 21 Apr 2007 03:12:35 -0000 --Sig_Jip+y.ksvo1mSrs3gahAUT8 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: quoted-printable Hello. I have tried sysutils/fusefs-ntfs (version 1.0) and had a maximum write speed of 1.2MB/Sec. Reading is a little faster: 2MB/Sec. There were some discussions about this in the ntfs-3g forums, and they said was fixed in the new beta version (now it's stable, see official site), note that by default it uses an option that is not available on FreeBSD's mount_fusefs, so try with "-o no_def_opts". http://forum.ntfs-3g.org/viewtopic.php?p=3D1330&sid=3D8e59dcb7050a15378eb93= d5659c04409 It makes no difference for me. However I found the following which says "the reason for the slow copy on FreeBSD is the lack of buffer cache for block devices which should be solved in FreeBSD 7.0", and mentions "ublio", a library for user space cache which can be used with it, made by the author of fuse4bsd. The rest of the thread is irrelevant for the matter. Is this change going to be MFC'ed to RELENG_6? http://forum.ntfs-3g.org/viewtopic.php?p=3D1153&sid=3Dcde9378447762e86345a8= 9130fd267d5 Unfortunately I couldn't make it work, if someone has time, please take a look. It would be really appreciated. I tried contacting the port maintainer and fuse4bsd/libublio author without response. Also I have posted this to freebsd-hackers@ without response, maybe this is the right place for this kind of discussions. Thanks and Best Regards, Ale P.S.: please CC me as I'm not subscribed. --Sig_Jip+y.ksvo1mSrs3gahAUT8 Content-Type: application/pgp-signature; name=signature.asc Content-Disposition: attachment; filename=signature.asc -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (FreeBSD) iD8DBQFGKX95iV05EpRcP2ERAms9AJ9bz2ioT/wPvKoXv1Dg+UXdVxSANACfciYC iuUt/L6Box5Ft4Jm3BbP+H8= =PLS4 -----END PGP SIGNATURE----- --Sig_Jip+y.ksvo1mSrs3gahAUT8-- From owner-freebsd-performance@FreeBSD.ORG Sat Apr 21 03:24:47 2007 Return-Path: X-Original-To: freebsd-performance@FreeBSD.org Delivered-To: freebsd-performance@FreeBSD.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id C558616A401 for ; Sat, 21 Apr 2007 03:24:47 +0000 (UTC) (envelope-from alepulver@FreeBSD.org) Received: from relay00.pair.com (relay00.pair.com [209.68.5.9]) by mx1.freebsd.org (Postfix) with SMTP id 6609213C44C for ; Sat, 21 Apr 2007 03:24:47 +0000 (UTC) (envelope-from alepulver@FreeBSD.org) Received: (qmail 28311 invoked by uid 0); 21 Apr 2007 02:58:06 -0000 Received: from 190.55.91.88 (HELO deimos.mars.bsd) (190.55.91.88) by relay00.pair.com with SMTP; 21 Apr 2007 02:58:06 -0000 X-pair-Authenticated: 190.55.91.88 Date: Fri, 20 Apr 2007 23:57:47 -0300 From: Alejandro Pulver To: freebsd-performance@FreeBSD.org Message-ID: <20070420235747.5b83996d@deimos.mars.bsd> X-Mailer: Claws Mail 2.8.1 (GTK+ 2.10.11; i386-portbld-freebsd6.2) Mime-Version: 1.0 Content-Type: multipart/signed; boundary=Sig_nx9FMI7yBEIq0INZdRL9xWZ; protocol="application/pgp-signature"; micalg=PGP-SHA1 X-Mailman-Approved-At: Sat, 21 Apr 2007 04:41:42 +0000 Cc: Subject: Gaim log writing delays the system 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: Sat, 21 Apr 2007 03:24:47 -0000 --Sig_nx9FMI7yBEIq0INZdRL9xWZ Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: quoted-printable Hello. I have enabled logging in Gaim, and when a chat message arrives and it is logged the disk writing delays (freezes) the system for less than a second, it can be noticed for example with XMMS which does a strange sound during that period. I think this problem is related to the system and not the port, that's why I asked here. Also I guess more information is needed about this, like ktrace/truss output of Gaim together with kernel statistics (vmstat/iostat). But other than ktrace, I don't use them. What would be the commands to get the most relevant information about this? I am using FreeBSD 6.2-RELEASE, and the boot message is here (the file dmesg_machine_2.txt): http://people.freebsd.org/~alepulver/disk-crash.tar.bz2 I posted this in freebsd-hackers@ and got no answer, maybe this is the right place. Thanks and Best Regards, Ale P.S.: please CC me as I'm not subscribed. --Sig_nx9FMI7yBEIq0INZdRL9xWZ Content-Type: application/pgp-signature; name=signature.asc Content-Disposition: attachment; filename=signature.asc -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (FreeBSD) iD8DBQFGKX2riV05EpRcP2ERAv9bAJ9rMud06Vle2GJyI9lZ8Bs3otPwYgCgwBzq MKYeXjn7bWBG3Ms7Xv85iCU= =+Tnr -----END PGP SIGNATURE----- --Sig_nx9FMI7yBEIq0INZdRL9xWZ-- From owner-freebsd-performance@FreeBSD.ORG Sat Apr 21 15:45:47 2007 Return-Path: X-Original-To: freebsd-performance@freebsd.org Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id C417616A400 for ; Sat, 21 Apr 2007 15:45:47 +0000 (UTC) (envelope-from cheffo@FreeBSD-BG.org) Received: from blah.sun-fish.com (blah.sun-fish.com [217.18.249.150]) by mx1.freebsd.org (Postfix) with ESMTP id 7E02113C46C for ; Sat, 21 Apr 2007 15:45:47 +0000 (UTC) (envelope-from cheffo@FreeBSD-BG.org) Received: from blah.sun-fish.com (localhost [127.0.0.1]) by blah.sun-fish.com (Postfix) with ESMTP id D3C631B10EA4 for ; Sat, 21 Apr 2007 17:45:45 +0200 (CEST) Received: from [192.168.3.125] (hater.cmotd.com [192.168.3.125]) by blah.sun-fish.com (Postfix) with ESMTP id D147A1B10C26 for ; Sat, 21 Apr 2007 17:45:45 +0200 (CEST) Message-ID: <462A31A9.5080207@FreeBSD-BG.org> Date: Sat, 21 Apr 2007 18:45:45 +0300 From: Cheffo User-Agent: Thunderbird 1.5.0.10 (X11/20070326) MIME-Version: 1.0 To: freebsd-performance@freebsd.org Content-Type: text/plain; charset=windows-1251; format=flowed Content-Transfer-Encoding: 7bit X-Virus-Scanned: ClamAV using ClamSMTP on BLAH Subject: apache httpd performance 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: Sat, 21 Apr 2007 15:45:47 -0000 Hello, I'm running simple benchmark, ab (freebsd) -> apache 1.3.37 (freebsd) and ab (freebsd) -> apache 1.3.34 (linux) I do not pretend that the benchmark is accurate :) and do not really care about results in linux, but they are here just for comparison. ab -t 30 -c 100 http://boar.cmotd.com:81/manual/ (boar is freebsd) Finished 26678 requests Server Software: Apache/1.3.37 Server Hostname: boar.cmotd.com Server Port: 81 Document Path: /manual/ Document Length: 9364 bytes Concurrency Level: 100 Time taken for tests: 30.000 seconds Complete requests: 26678 Failed requests: 0 Broken pipe errors: 0 Total transferred: 260937758 bytes HTML transferred: 249863898 bytes Requests per second: 889.27 [#/sec] (mean) Time per request: 112.45 [ms] (mean) Time per request: 1.12 [ms] (mean, across all concurrent requests) Transfer rate: 8697.93 [Kbytes/sec] received And this is what I have when the host is linux (some version of ubuntu not tunned just base install) ab -t 30 -c 100 http://dany.cmotd.com:80/manual/ Server Software: Apache/1.3.34 Server Hostname: dany.cmotd.com Server Port: 80 Document Path: /manual/ Document Length: 9158 bytes Concurrency Level: 100 Time taken for tests: 30.000 seconds Complete requests: 32848 Failed requests: 0 Broken pipe errors: 0 Total transferred: 309841821 bytes HTML transferred: 300932967 bytes Requests per second: 1094.93 [#/sec] (mean) Time per request: 91.33 [ms] (mean) Time per request: 0.91 [ms] (mean, across all concurrent requests) Transfer rate: 10328.06 [Kbytes/sec] received changing "-c" from 100 to 500 doesn't show big difference in final results. What can I do to improve my freebsd web server ? cat /etc/sysctl.conf kern.fallback_elf_brand=3 net.inet.ip.random_id=1 (I changed this 0, but nothing changed) kern.maxfiles=25000 net.inet.tcp.delayed_ack=0 net.inet.tcp.inflight.enable: 0 (0/1 does not make difference) CPU: AMD Athlon(tm) 64 Processor 3200+ (2002.57-MHz K8-class CPU) FreeBSD 6.2-STABLE amd64 both httpd.conf have: MinSpareServers 10 MaxSpareServers 20 MaxClients 500 MaxRequestsPerChild 0 StartServers 10 And yes I know "StartServers 10" is not optimal for benchmarking, but this is not the problem. I also found that increasing kern.ipc.maxsockets to 16384+ give a boost and "completed requests" jump to: 28067 but it is still not 32K :) Increasing kern.ipc.nmbclusters doesn't change the picture at all. What else I can change/test to improve performance? -- Best Wishes, Stefan Lambrev ICQ# 24134177 From owner-freebsd-performance@FreeBSD.ORG Sat Apr 21 23:10:57 2007 Return-Path: X-Original-To: freebsd-performance@freebsd.org Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 7F83216A401 for ; Sat, 21 Apr 2007 23:10:57 +0000 (UTC) (envelope-from gofp-freebsd-performance@m.gmane.org) Received: from ciao.gmane.org (main.gmane.org [80.91.229.2]) by mx1.freebsd.org (Postfix) with ESMTP id 10B2213C448 for ; Sat, 21 Apr 2007 23:10:56 +0000 (UTC) (envelope-from gofp-freebsd-performance@m.gmane.org) Received: from list by ciao.gmane.org with local (Exim 4.43) id 1HfOjB-0002Pt-UT for freebsd-performance@freebsd.org; Sun, 22 Apr 2007 01:10:49 +0200 Received: from 89-172-36-120.adsl.net.t-com.hr ([89.172.36.120]) by main.gmane.org with esmtp (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Sun, 22 Apr 2007 01:10:49 +0200 Received: from ivoras by 89-172-36-120.adsl.net.t-com.hr with local (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Sun, 22 Apr 2007 01:10:49 +0200 X-Injected-Via-Gmane: http://gmane.org/ To: freebsd-performance@freebsd.org From: Ivan Voras Date: Sun, 22 Apr 2007 01:10:35 +0200 Lines: 105 Message-ID: References: <462A31A9.5080207@FreeBSD-BG.org> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="------------enigC93FCD43AD7C96A493EF15E7" X-Complaints-To: usenet@sea.gmane.org X-Gmane-NNTP-Posting-Host: 89-172-36-120.adsl.net.t-com.hr User-Agent: Thunderbird 1.5.0.10 (Windows/20070221) In-Reply-To: <462A31A9.5080207@FreeBSD-BG.org> X-Enigmail-Version: 0.94.3.0 Sender: news Subject: Re: apache httpd performance 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: Sat, 21 Apr 2007 23:10:57 -0000 This is an OpenPGP/MIME signed message (RFC 2440 and 3156) --------------enigC93FCD43AD7C96A493EF15E7 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Cheffo wrote: > What else I can change/test to improve performance? First you'll have to give more info about the hardware on both systems, and the way you benchmarked them (e.g. did you benchmark over ethernet or from the same machine?). There are also a bunch of things that may make apache go faster/slower, for example DNS resolving for the logs, rewrite rules, etc. - you should test with (as much as they can be) identical configurations. On a slow-ish 2 CPU Pentium3 server, over gigabit network (remote client), I get: Server Software: Apache/2.2.3 Server Port: 80 Document Path: /file.txt Document Length: 9500 bytes Concurrency Level: 100 Time taken for tests: 27.300259 seconds Complete requests: 50000 Failed requests: 0 Write errors: 0 Total transferred: 489226808 bytes HTML transferred: 475025956 bytes Requests per second: 1831.48 [#/sec] (mean) Time per request: 54.601 [ms] (mean) Time per request: 0.546 [ms] (mean, across all concurrent requests)= Transfer rate: 17500.20 [Kbytes/sec] received On the same server, ab running locally, I get: Server Software: Apache/2.2.3 Server Port: 80 Document Path: /file.txt Document Length: 9500 bytes Concurrency Level: 100 Time taken for tests: 30.8557 seconds Complete requests: 29194 Failed requests: 6 (Connect: 6, Length: 0, Exceptions: 0) Write errors: 0 Total transferred: 286117856 bytes HTML transferred: 277811424 bytes Requests per second: 972.86 [#/sec] (mean) Time per request: 102.790 [ms] (mean) Time per request: 1.028 [ms] (mean, across all concurrent requests)= Transfer rate: 9311.04 [Kbytes/sec] received (Yes, this is a different version of apache than yours, but I'm illustrating a point :) ) Some more data points: - Remote client, apache, using keepalives: Requests per second: 2891.61 [#/sec] (mean) Time per request: 34.583 [ms] (mean) Time per request: 0.346 [ms] (mean, across all concurrent requests)= Transfer rate: 27729.00 [Kbytes/sec] received - Remote client, using thttpd instead of apache (keepalives have no influence here): Requests per second: 3728.68 [#/sec] (mean) Time per request: 26.819 [ms] (mean) Time per request: 0.268 [ms] (mean, across all concurrent requests)= Transfer rate: 35499.92 [Kbytes/sec] received Regarding this last one: It seems that FreeBSD really benefits from using 2 CPUs here. thttpd is a single-threaded async server, but the load on the machine shows cca 10% idle. Thttpd gets ~~45%, swi:net gets 45%, irq20:bge0 gets 20%, syslogd gets 5% and the rest goes where top can't follow. It looks like ipfw might be one of the limiting factors here (I use dynamic rules and the log shows ipfw discarding packets that look valid). --------------enigC93FCD43AD7C96A493EF15E7 Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.4 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFGKpnrldnAQVacBcgRAmueAKDtJ9KllCn8jYbAuDRSYC+IkSuzcQCghMsm gUtcsV9tulu8pyTdGKYpr2k= =MdsL -----END PGP SIGNATURE----- --------------enigC93FCD43AD7C96A493EF15E7--