From owner-freebsd-performance@FreeBSD.ORG Tue Jul 17 20:47:05 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 3873016A401 for ; Tue, 17 Jul 2007 20:47:05 +0000 (UTC) (envelope-from jack.toering@leadingedgeita.com) Received: from mail2.hspheredns.com (mail2.hspheredns.com [207.210.215.4]) by mx1.freebsd.org (Postfix) with SMTP id E1E9413C47E for ; Tue, 17 Jul 2007 20:47:04 +0000 (UTC) (envelope-from jack.toering@leadingedgeita.com) Received: (qmail 6635 invoked by uid 399); 17 Jul 2007 20:20:15 -0000 Received: from unknown (HELO DELL9400) (jack.toering@leadingedgeita.com@71.205.40.245) by mail2.hspheredns.com with ESMTP; 17 Jul 2007 20:20:15 -0000 X-Originating-IP: 71.205.40.245 From: "Jack Toering" To: Date: Tue, 17 Jul 2007 16:20:19 -0400 Organization: Leading Edge IT Architects Message-ID: <029401c7c8af$e5d19c60$6480010a@DELL9400> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Mailer: Microsoft Office Outlook 11 thread-index: AcfIr9/iAf73cR4KSp6u8z7Ox6jxtg== X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.3138 X-Mailman-Approved-At: Tue, 17 Jul 2007 21:39:20 +0000 Subject: Which SMP CPU for FreeBSD 6.2? 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 Jul 2007 20:47:05 -0000 It's been a couple years since I had to buy another dedicated server. Back then the AMD Opteron had it hands down. The trade press seems to be indicating the pendulum has swung the other way and AMD is in trouble, especially since the Barcelona miss. Also, over the past two years I've read how code is optimized more for the Intel. I'd like your opinion on which way to go on a new server by picking one for each category. For the same price I can get: $159.00 Range Dual Opteron 246 Core 2 Duo E6600 2.4GHz + $30 and get Core 2 Duo E6700 2.66GHz $209.00 Range Dual Opteron 252 +$10 Opteron 2214 2.2Ghz $239.00 Range Xeon 5130 2.0GHz Woodcrest Clovertown 5310 1.6Ghz $279.00 Range Woodcrest 5160 Dual Core Dual 3Ghz Opteron 2216 HE 2.4GHz Thanks! From owner-freebsd-performance@FreeBSD.ORG Tue Jul 17 22:05:38 2007 Return-Path: X-Original-To: freebsd-performance@freebsd.org Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 758E316A400 for ; Tue, 17 Jul 2007 22:05:38 +0000 (UTC) (envelope-from almarrie@gmail.com) Received: from qb-out-0506.google.com (qb-out-0506.google.com [72.14.204.224]) by mx1.freebsd.org (Postfix) with ESMTP id 2F44F13C4B4 for ; Tue, 17 Jul 2007 22:05:38 +0000 (UTC) (envelope-from almarrie@gmail.com) Received: by qb-out-0506.google.com with SMTP id a10so1525328qbd for ; Tue, 17 Jul 2007 15:05:37 -0700 (PDT) DKIM-Signature: a=rsa-sha1; c=relaxed/relaxed; d=gmail.com; s=beta; h=domainkey-signature:received:received:message-id:date:from:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=h/3wUvxZWgGvkfOYlYQTdk+z6Moj52dVnUvu6QGli18StEaRtJnjQ88DE0yV2fJ0ulcRmgYAHDYBNyez9BTOZaO7VXpFOoEt8odIgyziUR9Q4ccM3ZypMAguvlAhYJgGSBHXQ0Sa3YPjThAsQMBnoy5biRBxszz0dA0MIiB/VIo= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=received:message-id:date:from:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=DFpXp51ygzkqN+U8jo+jJjFG7eygpkA3Lgt5eMAxQouTXy4mKl7ZW54+R/VLmcUcUrEWMfqPRATOetoeZShTvrSLVN9FX04NkcAIkf02tW++D8tXAIZzRV84Kbndnv8i1I7FZ7cRLXVEUeqSmJg2B+gSHpvufy6AlrYyvbdsaaM= Received: by 10.100.152.9 with SMTP id z9mr405596and.1184709108147; Tue, 17 Jul 2007 14:51:48 -0700 (PDT) Received: by 10.100.9.14 with HTTP; Tue, 17 Jul 2007 14:51:48 -0700 (PDT) Message-ID: <499c70c0707171451h1c638275p7e5d9277ed084539@mail.gmail.com> Date: Wed, 18 Jul 2007 00:51:48 +0300 From: "Abdullah Ibn Hamad Al-Marri" To: "Jack Toering" In-Reply-To: <029401c7c8af$e5d19c60$6480010a@DELL9400> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline References: <029401c7c8af$e5d19c60$6480010a@DELL9400> Cc: freebsd-performance@freebsd.org Subject: Re: Which SMP CPU for FreeBSD 6.2? 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 Jul 2007 22:05:38 -0000 On 7/17/07, Jack Toering wrote: > It's been a couple years since I had to buy another dedicated server. Back > then the AMD Opteron had it hands down. The trade press seems to be > indicating the pendulum has swung the other way and AMD is in trouble, > especially since the Barcelona miss. Also, over the past two years I've > read how code is optimized more for the Intel. I'd like your opinion on > which way to go on a new server by picking one for each category. For the > same price I can get: > > $159.00 Range > Dual Opteron 246 > Core 2 Duo E6600 2.4GHz > + $30 and get Core 2 Duo E6700 2.66GHz > > $209.00 Range > Dual Opteron 252 > +$10 Opteron 2214 2.2Ghz > > $239.00 Range > Xeon 5130 2.0GHz Woodcrest > Clovertown 5310 1.6Ghz > > $279.00 Range > Woodcrest 5160 Dual Core Dual 3Ghz > Opteron 2216 HE 2.4GHz > > Thanks! I would suggest Intel C2D E6600 with 2 GB of RAM and fast hd ;) -- Regards, -Abdullah Ibn Hamad Al-Marri Arab Portal http://www.WeArab.Net/ From owner-freebsd-performance@FreeBSD.ORG Tue Jul 17 22:26:02 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 36D0916A400 for ; Tue, 17 Jul 2007 22:26:02 +0000 (UTC) (envelope-from kometen@gmail.com) Received: from wx-out-0506.google.com (wx-out-0506.google.com [66.249.82.236]) by mx1.freebsd.org (Postfix) with ESMTP id 965C413C441 for ; Tue, 17 Jul 2007 22:26:01 +0000 (UTC) (envelope-from kometen@gmail.com) Received: by wx-out-0506.google.com with SMTP id i29so1388224wxd for ; Tue, 17 Jul 2007 15:26:01 -0700 (PDT) DKIM-Signature: a=rsa-sha1; c=relaxed/relaxed; d=gmail.com; s=beta; h=domainkey-signature:received:received:message-id:date:from:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=BTzjLgODIFogSVi8ulxiEP53WTCmB8W7EQSpcxuvBumr6LRclUuuXPUp8TunXfIu2ntfjG8dM5CG3/3cW0CHlMDXxf3ue7nE0Cx0J1Nu2cvAH8Ii4rzHRG8jxR7fcqwDf7KY8rSEl/WMWD2AG9+M/JGeFAfwdtHmkHFKLobHc24= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=received:message-id:date:from:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=XHD5TpqTHg/CGIOg+Z6vouGBjjg2+i0FYdJYmsl8KiDC/Ph5ir4zul3HbA6KhrJWmh1LZLLEEaHr59hYFY34ZatUg6tzf2YgzQLHvF65MbT2VJSGNUOCcILRCnyR5PMk9fpvI138Hc4KOKtNBkXshu3ilIAjSlkui27VPUbEk00= Received: by 10.70.59.20 with SMTP id h20mr1415408wxa.1184709512724; Tue, 17 Jul 2007 14:58:32 -0700 (PDT) Received: by 10.70.41.4 with HTTP; Tue, 17 Jul 2007 14:58:32 -0700 (PDT) Message-ID: Date: Tue, 17 Jul 2007 23:58:32 +0200 From: "Claus Guttesen" To: "Jack Toering" In-Reply-To: <029401c7c8af$e5d19c60$6480010a@DELL9400> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline References: <029401c7c8af$e5d19c60$6480010a@DELL9400> Cc: freebsd-performance@freebsd.org Subject: Re: Which SMP CPU for FreeBSD 6.2? 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 Jul 2007 22:26:02 -0000 > It's been a couple years since I had to buy another dedicated server. Back > then the AMD Opteron had it hands down. The trade press seems to be > indicating the pendulum has swung the other way and AMD is in trouble, > especially since the Barcelona miss. Also, over the past two years I've > read how code is optimized more for the Intel. I'd like your opinion on > which way to go on a new server by picking one for each category. For the > same price I can get: > > $159.00 Range > Dual Opteron 246 > Core 2 Duo E6600 2.4GHz > + $30 and get Core 2 Duo E6700 2.66GHz > > $209.00 Range > Dual Opteron 252 > +$10 Opteron 2214 2.2Ghz > > $239.00 Range > Xeon 5130 2.0GHz Woodcrest > Clovertown 5310 1.6Ghz > > $279.00 Range > Woodcrest 5160 Dual Core Dual 3Ghz > Opteron 2216 HE 2.4GHz I'm very delighted with our four-way woodcrest at 3 Ghz. It's a HP DL 380 G5. I have two four-way opterons at 2 Ghz but hz for hz the woodcrest is (way) faster. -- regards Claus When lenity and cruelty play for a kingdom, the gentlest gamester is the soonest winner. Shakespeare From owner-freebsd-performance@FreeBSD.ORG Tue Jul 17 22:04:36 2007 Return-Path: X-Original-To: freebsd-performance@freebsd.org Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 1BE7B16A404 for ; Tue, 17 Jul 2007 22:04:36 +0000 (UTC) (envelope-from jack.toering@leadingedgeita.com) Received: from mail2.hspheredns.com (mail2.hspheredns.com [207.210.215.4]) by mx1.freebsd.org (Postfix) with SMTP id DC49513C4BC for ; Tue, 17 Jul 2007 22:04:35 +0000 (UTC) (envelope-from jack.toering@leadingedgeita.com) Received: (qmail 25066 invoked by uid 399); 17 Jul 2007 22:04:26 -0000 Received: from unknown (HELO DELL9400) (jack.toering@leadingedgeita.com@71.205.40.245) by mail2.hspheredns.com with ESMTP; 17 Jul 2007 22:04:26 -0000 X-Originating-IP: 71.205.40.245 From: "Jack Toering" To: References: <029401c7c8af$e5d19c60$6480010a@DELL9400> <499c70c0707171451h1c638275p7e5d9277ed084539@mail.gmail.com> Date: Tue, 17 Jul 2007 18:04:31 -0400 Organization: Leading Edge IT Architects Message-ID: <02b001c7c8be$7443a480$6480010a@DELL9400> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Mailer: Microsoft Office Outlook 11 thread-index: AcfIvKiEEmXyJnx0S2KIk8O+rxD7dQAAMXXA X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.3138 In-reply-to: <499c70c0707171451h1c638275p7e5d9277ed084539@mail.gmail.com> X-Mailman-Approved-At: Wed, 18 Jul 2007 00:08:23 +0000 Subject: RE: Which SMP CPU for FreeBSD 6.2? 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 Jul 2007 22:04:36 -0000 >> $159.00 Range >> Dual Opteron 246 >> Core 2 Duo E6600 2.4GHz >> + $30 and get Core 2 Duo E6700 2.66GHz >> >> $209.00 Range >> Dual Opteron 252 >> +$10 Opteron 2214 2.2Ghz >> >> $239.00 Range >> Xeon 5130 2.0GHz Woodcrest >> Clovertown 5310 1.6Ghz >> >> $279.00 Range >> Woodcrest 5160 Dual Core Dual 3Ghz >> Opteron 2216 HE 2.4GHz >I would suggest Intel C2D E6600 with 2 GB of RAM and fast hd ;)< People seem to be happy with that CPU. Thank you for your response! -----Original Message----- From: Abdullah Ibn Hamad Al-Marri [mailto:almarrie@gmail.com] Sent: Tuesday, July 17, 2007 5:52 PM To: Jack Toering Cc: freebsd-performance@freebsd.org Subject: Re: Which SMP CPU for FreeBSD 6.2? On 7/17/07, Jack Toering wrote: > It's been a couple years since I had to buy another dedicated server. > Back then the AMD Opteron had it hands down. The trade press seems to > be indicating the pendulum has swung the other way and AMD is in > trouble, especially since the Barcelona miss. Also, over the past two > years I've read how code is optimized more for the Intel. I'd like > your opinion on which way to go on a new server by picking one for > each category. For the same price I can get: > > $159.00 Range > Dual Opteron 246 > Core 2 Duo E6600 2.4GHz > + $30 and get Core 2 Duo E6700 2.66GHz > > $209.00 Range > Dual Opteron 252 > +$10 Opteron 2214 2.2Ghz > > $239.00 Range > Xeon 5130 2.0GHz Woodcrest > Clovertown 5310 1.6Ghz > > $279.00 Range > Woodcrest 5160 Dual Core Dual 3Ghz > Opteron 2216 HE 2.4GHz > > Thanks! I would suggest Intel C2D E6600 with 2 GB of RAM and fast hd ;) -- Regards, -Abdullah Ibn Hamad Al-Marri Arab Portal http://www.WeArab.Net/ From owner-freebsd-performance@FreeBSD.ORG Tue Jul 17 22:10:24 2007 Return-Path: X-Original-To: freebsd-performance@freebsd.org Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 8862316A400 for ; Tue, 17 Jul 2007 22:10:24 +0000 (UTC) (envelope-from jack.toering@leadingedgeita.com) Received: from mail2.hspheredns.com (mail2.hspheredns.com [207.210.215.4]) by mx1.freebsd.org (Postfix) with SMTP id 53D5713C48E for ; Tue, 17 Jul 2007 22:10:24 +0000 (UTC) (envelope-from jack.toering@leadingedgeita.com) Received: (qmail 315 invoked by uid 399); 17 Jul 2007 22:10:15 -0000 Received: from unknown (HELO DELL9400) (jack.toering@leadingedgeita.com@71.205.40.245) by mail2.hspheredns.com with ESMTP; 17 Jul 2007 22:10:15 -0000 X-Originating-IP: 71.205.40.245 From: "Jack Toering" To: References: <029401c7c8af$e5d19c60$6480010a@DELL9400> Date: Tue, 17 Jul 2007 18:10:20 -0400 Organization: Leading Edge IT Architects Message-ID: <02b101c7c8bf$442c2140$6480010a@DELL9400> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Mailer: Microsoft Office Outlook 11 thread-index: AcfIvZqD3qcfXzX/TqCOj65Mz8kwEgAAHs0g X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.3138 In-reply-to: X-Mailman-Approved-At: Wed, 18 Jul 2007 00:08:23 +0000 Subject: RE: Which SMP CPU for FreeBSD 6.2? 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 Jul 2007 22:10:24 -0000 >I'm very delighted with our four-way woodcrest at 3 Ghz. It's a HP DL 380 G5. I have two four-way opterons at 2 Ghz but hz for hz the woodcrest is (way) faster.< These are things I need to hear bacause it doesn't make sense for me to watch these things until I'm in the market because the technology moves so fast. Thank you very much for your response! -----Original Message----- From: Claus Guttesen [mailto:kometen@gmail.com] Sent: Tuesday, July 17, 2007 5:59 PM To: Jack Toering Cc: freebsd-performance@freebsd.org Subject: Re: Which SMP CPU for FreeBSD 6.2? > It's been a couple years since I had to buy another dedicated server. > Back then the AMD Opteron had it hands down. The trade press seems to > be indicating the pendulum has swung the other way and AMD is in > trouble, especially since the Barcelona miss. Also, over the past two > years I've read how code is optimized more for the Intel. I'd like > your opinion on which way to go on a new server by picking one for > each category. For the same price I can get: > > $159.00 Range > Dual Opteron 246 > Core 2 Duo E6600 2.4GHz > + $30 and get Core 2 Duo E6700 2.66GHz > > $209.00 Range > Dual Opteron 252 > +$10 Opteron 2214 2.2Ghz > > $239.00 Range > Xeon 5130 2.0GHz Woodcrest > Clovertown 5310 1.6Ghz > > $279.00 Range > Woodcrest 5160 Dual Core Dual 3Ghz > Opteron 2216 HE 2.4GHz I'm very delighted with our four-way woodcrest at 3 Ghz. It's a HP DL 380 G5. I have two four-way opterons at 2 Ghz but hz for hz the woodcrest is (way) faster. -- regards Claus When lenity and cruelty play for a kingdom, the gentlest gamester is the soonest winner. Shakespeare From owner-freebsd-performance@FreeBSD.ORG Wed Jul 18 02:47:53 2007 Return-Path: X-Original-To: freebsd-performance@freebsd.org Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 5C8E716A401 for ; Wed, 18 Jul 2007 02:47:53 +0000 (UTC) (envelope-from anderson@freebsd.org) Received: from ns.trinitel.com (186.161.36.72.static.reverse.layeredtech.com [72.36.161.186]) by mx1.freebsd.org (Postfix) with ESMTP id 3192F13C4A8 for ; Wed, 18 Jul 2007 02:47:53 +0000 (UTC) (envelope-from anderson@freebsd.org) Received: from neutrino.vnode.org (r74-193-81-203.pfvlcmta01.grtntx.tl.dh.suddenlink.net [74.193.81.203]) (authenticated bits=0) by ns.trinitel.com (8.14.1/8.14.1) with ESMTP id l6I2T3ZR030890 (version=TLSv1/SSLv3 cipher=DHE-DSS-AES256-SHA bits=256 verify=NO); Tue, 17 Jul 2007 21:29:04 -0500 (CDT) (envelope-from anderson@freebsd.org) Message-ID: <469D7AF1.1090505@freebsd.org> Date: Tue, 17 Jul 2007 21:29:05 -0500 From: Eric Anderson User-Agent: Thunderbird 2.0.0.4 (X11/20070629) MIME-Version: 1.0 To: Jack Toering References: <029401c7c8af$e5d19c60$6480010a@DELL9400> <02b101c7c8bf$442c2140$6480010a@DELL9400> In-Reply-To: <02b101c7c8bf$442c2140$6480010a@DELL9400> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Spam-Status: No, score=0.0 required=5.0 tests=none autolearn=ham version=3.1.8 X-Spam-Checker-Version: SpamAssassin 3.1.8 (2007-02-13) on ns.trinitel.com Cc: freebsd-performance@freebsd.org Subject: Re: Which SMP CPU for FreeBSD 6.2? 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 Jul 2007 02:47:53 -0000 On 07/17/07 17:10, Jack Toering wrote: > >I'm very delighted with our four-way woodcrest at 3 Ghz. It's a HP DL 380 > G5. I have two four-way opterons at 2 Ghz but hz for hz the woodcrest is > (way) faster.< > > These are things I need to hear bacause it doesn't make sense for me to > watch these things until I'm in the market because the technology moves so > fast. The woodcrest is smokin' fast. The Core 2 Duo's (EXXXX numbers) are also very fast, so you won't be unhappy. In fact, the E6700 came out faster than woodcrest 5160 in a few tests. Eric > -----Original Message----- > From: Claus Guttesen [mailto:kometen@gmail.com] > Sent: Tuesday, July 17, 2007 5:59 PM > To: Jack Toering > Cc: freebsd-performance@freebsd.org > Subject: Re: Which SMP CPU for FreeBSD 6.2? > >> It's been a couple years since I had to buy another dedicated server. >> Back then the AMD Opteron had it hands down. The trade press seems to >> be indicating the pendulum has swung the other way and AMD is in >> trouble, especially since the Barcelona miss. Also, over the past two >> years I've read how code is optimized more for the Intel. I'd like >> your opinion on which way to go on a new server by picking one for >> each category. For the same price I can get: >> >> $159.00 Range >> Dual Opteron 246 >> Core 2 Duo E6600 2.4GHz >> + $30 and get Core 2 Duo E6700 2.66GHz >> >> $209.00 Range >> Dual Opteron 252 >> +$10 Opteron 2214 2.2Ghz >> >> $239.00 Range >> Xeon 5130 2.0GHz Woodcrest >> Clovertown 5310 1.6Ghz >> >> $279.00 Range >> Woodcrest 5160 Dual Core Dual 3Ghz >> Opteron 2216 HE 2.4GHz > > I'm very delighted with our four-way woodcrest at 3 Ghz. It's a HP DL 380 > G5. I have two four-way opterons at 2 Ghz but hz for hz the woodcrest is > (way) faster. > > -- > regards > Claus > > When lenity and cruelty play for a kingdom, the gentlest gamester is the > soonest winner. > > Shakespeare > > _______________________________________________ > freebsd-performance@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-performance > To unsubscribe, send any mail to "freebsd-performance-unsubscribe@freebsd.org" From owner-freebsd-performance@FreeBSD.ORG Wed Jul 18 12:19:30 2007 Return-Path: X-Original-To: freebsd-performance@freebsd.org Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 1BE1116A403 for ; Wed, 18 Jul 2007 12:19:30 +0000 (UTC) (envelope-from tom.hurst@clara.net) Received: from spork.qfe3.net (spork.qfe3.net [212.13.207.101]) by mx1.freebsd.org (Postfix) with ESMTP id D3D8313C471 for ; Wed, 18 Jul 2007 12:19:29 +0000 (UTC) (envelope-from tom.hurst@clara.net) Received: from [81.104.144.87] (helo=voi.aagh.net) by spork.qfe3.net with esmtp (Exim 4.66 (FreeBSD)) (envelope-from ) id 1IB8FD-000Jik-0c; Wed, 18 Jul 2007 13:03:03 +0100 Received: from freaky by voi.aagh.net with local (Exim 4.67 (FreeBSD)) (envelope-from ) id 1IB8FC-0003Ou-Oe; Wed, 18 Jul 2007 13:03:02 +0100 Date: Wed, 18 Jul 2007 13:03:02 +0100 From: Thomas Hurst To: Jack Toering Message-ID: <20070718120302.GA11968@voi.aagh.net> Mail-Followup-To: Jack Toering , freebsd-performance@freebsd.org References: <029401c7c8af$e5d19c60$6480010a@DELL9400> <02b101c7c8bf$442c2140$6480010a@DELL9400> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <02b101c7c8bf$442c2140$6480010a@DELL9400> Organization: Not much. User-Agent: Mutt/1.5.15 (2007-04-06) Sender: Thomas Hurst Cc: freebsd-performance@freebsd.org Subject: Re: Which SMP CPU for FreeBSD 6.2? 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 Jul 2007 12:19:30 -0000 * Jack Toering (Jack.Toering@LeadingEdgeITA.com) wrote: > These are things I need to hear bacause it doesn't make sense for me to > watch these things until I'm in the market because the technology moves so > fast. > > Thank you very much for your response! You probably don't need to follow every move, but it might make sense to do some actual research if performance matters to you? You don't even say what your performance critical server is going to be doing ;) Core 2 generally performs well, but K8's are still pretty good with many workloads thanks to their far superior interconnects. Most numbers I've seen have been more based on games and media encoding in single socket configurations, and in 32bit mode at that; not exactly interesting use cases for most servers. Things get even less clear cut when you start worrying about power consumption. One of our more CPU heavy server uses actually depends very heavily on memory bandwidth, which Intel are still lagging behind on quite significantly, and of course things like lights out management and driver support are generally far more critical than a 10% performance difference in $random_microbenchmark. -- Thomas 'Freaky' Hurst http://hur.st/ From owner-freebsd-performance@FreeBSD.ORG Wed Jul 18 13:05:04 2007 Return-Path: X-Original-To: freebsd-performance@freebsd.org Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id BDFEB16A40F for ; Wed, 18 Jul 2007 13:05:04 +0000 (UTC) (envelope-from jack.toering@leadingedgeita.com) Received: from mail2.hspheredns.com (mail2.hspheredns.com [207.210.215.4]) by mx1.freebsd.org (Postfix) with SMTP id 85B3613C4B5 for ; Wed, 18 Jul 2007 13:05:04 +0000 (UTC) (envelope-from jack.toering@leadingedgeita.com) Received: (qmail 2512 invoked by uid 399); 18 Jul 2007 13:04:59 -0000 Received: from unknown (HELO DELL9400) (jack.toering@leadingedgeita.com@71.205.40.245) by mail2.hspheredns.com with ESMTP; 18 Jul 2007 13:04:59 -0000 X-Originating-IP: 71.205.40.245 From: "Jack Toering" To: References: <029401c7c8af$e5d19c60$6480010a@DELL9400> <02b101c7c8bf$442c2140$6480010a@DELL9400> <20070718120302.GA11968@voi.aagh.net> Date: Wed, 18 Jul 2007 09:05:00 -0400 Organization: Leading Edge IT Architects Message-ID: <032101c7c93c$3fe17770$6480010a@DELL9400> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Mailer: Microsoft Office Outlook 11 thread-index: AcfJM5X5Uk1g0lMnTS2irnRc9a4AgwAAc/dA X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.3138 In-reply-to: <20070718120302.GA11968@voi.aagh.net> X-Mailman-Approved-At: Wed, 18 Jul 2007 13:06:34 +0000 Subject: RE: Which SMP CPU for FreeBSD 6.2? 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 Jul 2007 13:05:04 -0000 >You probably don't need to follow every move, but it might make sense to do some actual research if performance matters to you? You don't even say what your performance critical server is going to be doing ;)< At the moment I'm not too sure either. :) The reason I say this is we have a 246 Dual Opteron that is handling 15,000 unique visitors an hour pulling data from around the world, generating weather maps, and pounding MySQL. It has no problems. There is host a large hotel site on the same server that is doing 900 and hour. 2/3 of the bandwidth is inbound to make the calculations. It took us a lot off tweaking to get it where it is today. >but K8's are still pretty good with many workloads thanks to their far superior interconnects.< Which is why we went with dual 246s at the time. The second processor on the Intel was a waste of money and it dual chip at the server level couldn't hold a candle to their laptop chips and the cost more than AMD. That decision was a no-brainer. >Most numbers I've seen have been more based on games and media encoding in single socket configurations, and in 32bit mode at that;< Yes, I'm quite skeptical about benchmarks even on web servers having learned the hard way. We tried 3 big names on Linux where the site fell down before 9AM. A 4th, Rackspace, the salesman didn't return my calls after they had an engineer check out what we were doing. Before and after we tried 3 versions off Linux ourselves with lots of configurations. Based on all of the benchmarks, FreeBSD didn't make sense which is why it took us so long to get here. Our competitors are running 5 box server farms to do what we accomplish with this setup and we have better response times. We are running SCSI raid. There is no measurable user response difference between a load of 17 or .7. Due to optimizations, our current load doesn't exceed 6 or 7 with normal being more around 2. Then the NOCs hated us because we used the bandwidth they promised and we were told to move along one way or another 3 times. The best NOC we've ever had is SAVVIS, Texas. They could not handle it either where the server was at first, but after I proved to them with a packet sniffer that there was a problem, they worked hard for us to get us a place in the DC that could handle it. We've lived happily ever after there. That's where the next one is going also. We've had all we can handle of being a fugitive and a vagabond. Naturally, my thoughts are drifting toward, "Hey, the new site should be doing the 246s and the old site something new." In the mean time I've kept up off and on, on the issues with AMD code not being nearly as efficient as the Intel code due to optimization for the Intel due to its far greater numbers. Next, the abosolute 180 degree change in the market between now and then concerning AMD and Intel. I've read the Tom's hardware and AnandTech (I think) comparisions. However, a lot of beautify theories get beat up by ugly realities and I have the scars to prove it. That's why I'm here fishing. I'm not expecting to find someone else with traffic and loads like this, but any insights are welcome when it comes to rolling the dice again because anything you can do to cut down the number of tries you need to throw the dice is a huge help. Thanks! From owner-freebsd-performance@FreeBSD.ORG Wed Jul 18 21:59:06 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 CCF3F16A403 for ; Wed, 18 Jul 2007 21:59:06 +0000 (UTC) (envelope-from darcy@ok-connect.com) Received: from dbitech.homenet.dbitech.bc.ca (S010600a08330057a.ok.shawcable.net [24.67.72.171]) by mx1.freebsd.org (Postfix) with ESMTP id 8A91B13C471 for ; Wed, 18 Jul 2007 21:59:06 +0000 (UTC) (envelope-from darcy@ok-connect.com) Received: from dbitech.homenet.dbitech.bc.ca (localhost [127.0.0.1]) by dbitech.homenet.dbitech.bc.ca (8.13.8/8.13.8) with ESMTP id l6IKiciY054863; Wed, 18 Jul 2007 13:44:38 -0700 (PDT) (envelope-from darcy@ok-connect.com) Received: from localhost (localhost [[UNIX: localhost]]) by dbitech.homenet.dbitech.bc.ca (8.13.8/8.13.8/Submit) id l6IKiWMb054862; Wed, 18 Jul 2007 13:44:32 -0700 (PDT) (envelope-from darcy@ok-connect.com) X-Authentication-Warning: dbitech.homenet.dbitech.bc.ca: darcy set sender to darcy@ok-connect.com using -f From: Darcy Buskermolen Organization: OK-connect.com To: freebsd-performance@freebsd.org Date: Wed, 18 Jul 2007 13:44:31 -0700 User-Agent: KMail/1.9.6 References: <029401c7c8af$e5d19c60$6480010a@DELL9400> <20070718120302.GA11968@voi.aagh.net> <032101c7c93c$3fe17770$6480010a@DELL9400> In-Reply-To: <032101c7c93c$3fe17770$6480010a@DELL9400> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200707181344.32037.darcy@ok-connect.com> X-Mailman-Approved-At: Wed, 18 Jul 2007 22:09:58 +0000 Cc: Jack Toering Subject: Re: Which SMP CPU for FreeBSD 6.2? 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 Jul 2007 21:59:06 -0000 On Wednesday 18 July 2007 06:05:00 Jack Toering wrote: > >You probably don't need to follow every move, but it might make sense to > > do > > some actual research if performance matters to you? You don't even say > what your performance critical server is going to be doing ;)< > > At the moment I'm not too sure either. :) The reason I say this is we have > a 246 Dual Opteron that is handling 15,000 unique visitors an hour pulling > data from around the world, generating weather maps, and pounding MySQL. > It has no problems. There is host a large hotel site on the same server > that is doing 900 and hour. 2/3 of the bandwidth is inbound to make the > calculations. It took us a lot off tweaking to get it where it is today. > > >but K8's are still pretty good with many workloads thanks to their far > > superior interconnects.< > > Which is why we went with dual 246s at the time. The second processor on > the Intel was a waste of money and it dual chip at the server level > couldn't hold a candle to their laptop chips and the cost more than AMD. > That decision was a no-brainer. > > >Most numbers I've seen have been more based on games and media encoding in > > single socket configurations, and in 32bit mode at that;< > > Yes, I'm quite skeptical about benchmarks even on web servers having > learned the hard way. We tried 3 big names on Linux where the site fell > down before 9AM. A 4th, Rackspace, the salesman didn't return my calls > after they had an engineer check out what we were doing. Before and after > we tried 3 versions off Linux ourselves with lots of configurations. Based > on all of the benchmarks, FreeBSD didn't make sense which is why it took us > so long to get here. Our competitors are running 5 box server farms to do > what we accomplish with this setup and we have better response times. We > are running SCSI raid. There is no measurable user response difference > between a load of 17 or .7. Due to optimizations, our current load doesn't > exceed 6 or 7 with normal being more around 2. > > Then the NOCs hated us because we used the bandwidth they promised and we > were told to move along one way or another 3 times. The best NOC we've > ever had is SAVVIS, Texas. They could not handle it either where the > server was at first, but after I proved to them with a packet sniffer that > there was a problem, they worked hard for us to get us a place in the DC > that could handle it. We've lived happily ever after there. That's where > the next one is going also. We've had all we can handle of being a > fugitive and a vagabond. > > Naturally, my thoughts are drifting toward, "Hey, the new site should be > doing the 246s and the old site something new." In the mean time I've kept > up off and on, on the issues with AMD code not being nearly as efficient as > the Intel code due to optimization for the Intel due to its far greater > numbers. Next, the abosolute 180 degree change in the market between now > and then concerning AMD and Intel. I've read the Tom's hardware and > AnandTech (I think) comparisions. However, a lot of beautify theories get > beat up by ugly realities and I have the scars to prove it. That's why I'm > here fishing. I'm not expecting to find someone else with traffic and > loads like this, but any insights are welcome when it comes to rolling the > dice again because anything you can do to cut down the number of tries you > need to throw the dice is a huge help. Another site you can look at for reviews which is far less game/windows/office oriented is tweakers.net, http://tweakers.net/reviews/661/7 for a good test of a woodcrest based server. > > Thanks! > > _______________________________________________ > 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" From owner-freebsd-performance@FreeBSD.ORG Wed Jul 18 22:40:05 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 7DCB316A402 for ; Wed, 18 Jul 2007 22:40:05 +0000 (UTC) (envelope-from x0dapara@gmail.com) Received: from nz-out-0506.google.com (nz-out-0506.google.com [64.233.162.228]) by mx1.freebsd.org (Postfix) with ESMTP id 2F44813C4B6 for ; Wed, 18 Jul 2007 22:40:05 +0000 (UTC) (envelope-from x0dapara@gmail.com) Received: by nz-out-0506.google.com with SMTP id l8so283019nzf for ; Wed, 18 Jul 2007 15:40:04 -0700 (PDT) DKIM-Signature: a=rsa-sha1; c=relaxed/relaxed; d=gmail.com; s=beta; h=domainkey-signature:received:received:message-id:date:from:to:subject:mime-version:content-type; b=TopHEri9Z6j4UTAuzsks/tJ2G21eG4lPSPp+fGr8RyYJ1lGnt5uS4FCquFs3p9Tsfhwe+u+WkVFcVxDIc8/bjydgsgBAUCpF8cYaDtZAO4wWS8OTzIyDsU+CCkECJv5/SeBJdxiUKxQgqkTR/l0yvG0gtXNYNHnxNK+SXLQIyso= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=received:message-id:date:from:to:subject:mime-version:content-type; b=dUBxkEKNORz5vuaU4CaGxdaPiyjcwFRA+fzV0Cgm9kD4sfJu8glUByNTZmZBq70du+2hgfSRlMRRst39gxg7IhiJ49N54lDW/X+szJuQSHYwR0mOTBd6jKGQIhIejWcze5tQAV6kh2KgFQ/W2UXpNZCVJmOVAAO8QBo+uxfGDGI= Received: by 10.142.103.6 with SMTP id a6mr159973wfc.1184796778243; Wed, 18 Jul 2007 15:12:58 -0700 (PDT) Received: by 10.143.44.2 with HTTP; Wed, 18 Jul 2007 15:12:57 -0700 (PDT) Message-ID: <2f0146460707181512x3841af57l588e4d6e67bd5884@mail.gmail.com> Date: Wed, 18 Jul 2007 23:12:57 +0100 From: "Michael Vaughn" To: freebsd-questions@freebsd.org, freebsd-performance@freebsd.org, freebsd-hackers@freebsd.org MIME-Version: 1.0 X-Mailman-Approved-At: Wed, 18 Jul 2007 23:44:01 +0000 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline X-Content-Filtered-By: Mailman/MimeDel 2.1.5 Cc: Subject: FreeBSD 6.2-STABLE && apache 2.2.4 = bad performance. Help! 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 Jul 2007 22:40:05 -0000 Hello everyone, I am contacting -performance, -questions, and -hackers in the hope someone helps me troubleshoot a problem with FreeBSD 6.2 and apache 2.2.4 uname: FreeBSD 6.2-STABLE Fri Jun 22 12:17:03 UTC 2007 amd64 installed php modules: php5-5.2.3 PHP Scripting Language (Apache Module and CLI) php5-gd-5.2.3 The gd shared extension for php php5-mysql-5.2.3 The mysql shared extension for php php5-pcre-5.2.3 The pcre shared extension for php php5-session-5.2.3 The session shared extension for php php5-simplexml-5.2.3 The simplexml shared extension for php php5-tokenizer-5.2.3 The tokenizer shared extension for php php5-xml-5.2.3 The xml shared extension for php apache version: apache-2.2.4_2 Version 2.2 of Apache web server with prefork MPM. system: real memory = 5100273664 (4864 MB) avail memory = 4120178688 (3929 MB) CPU: Intel(R) Xeon(TM) CPU 2.66GHz (2666.78-MHz K8-class CPU) Logical CPUs per core: 2 FreeBSD/SMP: Multiprocessor System Detected: 8 CPUs The problem: Right after starting apache, the loads on the server will climb to 10-40's and the application will become unacceptably slow. This will go on until few users are using the said application. (note: other servers running older FreeBSD versions on dual cpus running the same code don't exhibit this system% problem) top shows more than 60% of the CPU time is spent on system: CPU states: 19.9% user, 0.0% nice, 73.7% system, 1.7% interrupt, 4.7% idle Mem: 398M Active, 2226M Inact, 253M Wired, 202M Cache, 214M Buf, 567M Free The apache processes look like: PID USERNAME THR PRI NICE SIZE RES STATE C TIME WCPU COMMAND 56882 www 1 103 0 139M 17516K select 0 0:03 12.66% httpd 56862 www 1 100 0 139M 21168K CPU2 6 0:06 11.87% httpd 56830 www 1 99 0 138M 19684K select 2 0:09 10.76% httpd 56887 www 1 105 0 139M 17488K select 6 0:01 10.49% httpd 56852 www 1 99 0 138M 20352K select 4 0:06 10.26% httpd 56889 www 1 106 0 139M 17548K select 6 0:01 10.04% httpd 56894 www 1 109 0 139M 17024K select 6 0:01 9.79% httpd 56839 www 1 99 0 138M 21216K select 6 0:06 9.36% httpd 56866 www 1 99 0 138M 17664K select 6 0:04 9.36% httpd 56890 www 1 108 0 138M 16180K select 4 0:01 9.29% httpd 56848 www 1 99 0 138M 20460K select 2 0:06 9.27% httpd 56865 www 1 99 0 138M 18920K select 2 0:05 9.23% httpd 56883 www 1 102 0 138M 16744K select 4 0:02 8.99% httpd 56870 www 1 100 0 139M 18440K select 2 0:03 8.86% httpd 56850 www 1 98 0 138M 21284K select 6 0:05 8.84% httpd 56860 www 1 99 0 138M 19584K select 0 0:05 8.70% httpd 56864 www 1 99 0 139M 18028K select 2 0:04 8.23% httpd 56854 www 1 99 0 138M 20696K select 6 0:05 8.23% httpd 56853 www 1 98 0 138M 19564K select 4 0:06 8.11% httpd 56835 www 1 98 0 139M 20276K CPU6 4 0:07 8.10% httpd 56849 www 1 98 0 138M 19532K select 0 0:05 7.95% httpd 56851 www 1 98 0 139M 20252K select 4 0:05 7.35% httpd 56888 www 1 4 0 139M 17100K sbwait 6 0:01 7.31% httpd 56869 www 1 100 0 139M 18632K select 4 0:02 6.75% httpd 56861 www 1 98 0 139M 18404K select 0 0:04 6.58% httpd 56863 www 1 98 0 139M 20220K select 2 0:03 6.40% httpd 56867 www 1 99 0 138M 17452K select 6 0:03 6.39% httpd 56868 www 1 99 0 138M 18376K select 0 0:03 6.20% httpd 56893 www 1 107 0 138M 12964K select 0 0:00 5.62% httpd 56878 www 1 100 0 138M 16732K select 6 0:02 5.27% httpd 56881 www 1 100 0 138M 16288K select 6 0:01 2.18% httpd I had to lower MaxClients on apache substancially from 128 to 32, or loads would quickly go to 40+. (Other servers with dual cpus instead of quad and apache 1.3 on freebsd 6.0 don't have this problem) vmstat 1: procs memory page disks faults cpu r b w avm fre flt re pi po fr sr ad4 ad6 in sy cs us sy id 0 1 0 1380860 787212 1365 0 0 0 1312 1 0 0 486 559 842 13 22 65 1 1 0 1384588 787128 2724 0 0 0 2581 0 0 88 3038 82956 48776 19 38 43 4 1 0 1399232 782936 3328 0 0 0 2112 0 0 97 3592 101093 66497 24 50 26 0 1 2 1400200 781628 3726 0 0 0 2910 0 0 99 3529 100289 81531 23 58 19 19 1 0 1404000 778556 2263 0 0 0 1141 0 0 62 2964 73572 101432 19 76 5 15 1 1 1402452 776800 2499 0 0 0 1714 0 7 74 2965 68441 102276 19 78 3 15 1 0 1401548 777112 2213 0 0 0 2103 0 0 42 2491 105584 109418 15 79 6 8 1 1 1403324 778856 2606 0 0 0 2748 0 0 84 2996 75288 91676 22 76 2 0 1 3 1396864 781344 2764 0 0 0 3010 0 0 86 3393 90765 85952 25 70 5 1 2 0 1395520 782604 2774 0 0 0 2978 0 0 79 3195 88251 92623 20 63 17 6 1 0 1396096 781832 2641 0 0 0 2195 0 1 82 3347 96322 55942 21 42 37 iostat 1: tty ad4 ad6 ad8 cpu tin tout KB/t tps MB/s KB/t tps MB/s KB/t tps MB/s us ni sy in id 0 28 13.94 4 0.06 16.13 48 0.75 13.94 4 0.06 13 0 21 1 65 0 231 0.00 0 0.00 16.00 68 1.06 0.00 0 0.00 19 0 74 1 5 0 77 0.00 0 0.00 16.00 90 1.40 0.00 0 0.00 17 0 77 2 4 0 77 0.50 1 0.00 16.00 46 0.72 0.50 1 0.00 14 0 82 1 4 0 77 0.00 0 0.00 16.00 83 1.30 0.00 0 0.00 21 0 65 2 12 0 77 0.00 0 0.00 16.00 37 0.58 0.00 0 0.00 18 0 76 1 5 0 77 0.00 0 0.00 16.00 82 1.28 0.00 0 0.00 20 0 74 2 4 0 77 0.00 0 0.00 16.00 68 1.06 0.00 0 0.00 21 0 47 2 30 0 77 0.00 0 0.00 16.00 61 0.95 0.00 0 0.00 20 0 33 1 46 0 77 0.00 0 0.00 16.00 71 1.11 0.00 0 0.00 21 0 44 2 33 0 77 0.00 0 0.00 16.00 62 0.97 0.00 0 0.00 22 0 46 2 30 The kernel is custom built, with the following added (relevant) options: options PMAP_SHPGPERPROC=4096 options ACCEPT_FILTER_HTTP /boot/loader.conf: kern.maxproc=32768 kern.ipc.shmmni=4096 kern.ipc.shmseg=2048 kern.ipc.semmns=960 kern.ipc.semmni=160 kern.ipc.semume=160 kern.ipc.semmnu=480 mysqld is also running on this server, altough it never goes past 8-10% CPU usage. It is separated from apache on I/O, via jails and gmirror: mirror/gm0 COMPLETE ad4 ad8 mirror/gm1 COMPLETE ad6 ad10 Now this web application isn't the best code out there, but this is a quad cpu server and it's performing a lot worse than some servers I have running with 6.0 with apache 1.3 for over 400 days. Am I the only one getting terrible performance with apache2 on FreeBSD 6 ? p.s: cc me, I am not subscribed to any of the lists. Regards, Mark From owner-freebsd-performance@FreeBSD.ORG Thu Jul 19 02:33:43 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 DD73116A403 for ; Thu, 19 Jul 2007 02:33:43 +0000 (UTC) (envelope-from jack.toering@leadingedgeita.com) Received: from mail2.hspheredns.com (mail2.hspheredns.com [207.210.215.4]) by mx1.freebsd.org (Postfix) with SMTP id 8D4FC13C4B7 for ; Thu, 19 Jul 2007 02:33:43 +0000 (UTC) (envelope-from jack.toering@leadingedgeita.com) Received: (qmail 8444 invoked by uid 399); 19 Jul 2007 02:33:31 -0000 Received: from unknown (HELO DELL9400) (jack.toering@leadingedgeita.com@71.205.40.245) by mail2.hspheredns.com with ESMTP; 19 Jul 2007 02:33:31 -0000 X-Originating-IP: 71.205.40.245 From: "Jack Toering" To: References: <029401c7c8af$e5d19c60$6480010a@DELL9400> <20070718120302.GA11968@voi.aagh.net> <032101c7c93c$3fe17770$6480010a@DELL9400> <200707181344.32037.darcy@ok-connect.com> Date: Wed, 18 Jul 2007 22:33:38 -0400 Organization: Leading Edge IT Architects Message-ID: <003001c7c9ad$36e8a550$6480010a@DELL9400> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Mailer: Microsoft Office Outlook 11 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.3138 thread-index: AcfJfOK13a4srifyQyeIjST/CO7fzAAL5FPg In-Reply-To: <200707181344.32037.darcy@ok-connect.com> X-Mailman-Approved-At: Thu, 19 Jul 2007 03:51:42 +0000 Subject: RE: Which SMP CPU for FreeBSD 6.2? X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 19 Jul 2007 02:33:44 -0000 >Another site you can look at for reviews which is far less game/windows/office oriented is tweakers.net, http://tweakers.net/reviews/661/7 for a good test of a woodcrest based server.< Actually that is on the Clovertown, but it's a good read. I'll check out the others. That is more web server focused which I need to round things out. Great site. Thanks! From owner-freebsd-performance@FreeBSD.ORG Thu Jul 19 14:34:23 2007 Return-Path: X-Original-To: freebsd-performance@freebsd.org Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 96CEE16A402 for ; Thu, 19 Jul 2007 14:34:23 +0000 (UTC) (envelope-from jack.toering@leadingedgeita.com) Received: from mail2.hspheredns.com (mail2.hspheredns.com [207.210.215.4]) by mx1.freebsd.org (Postfix) with SMTP id 6262213C461 for ; Thu, 19 Jul 2007 14:34:23 +0000 (UTC) (envelope-from jack.toering@leadingedgeita.com) Received: (qmail 22185 invoked by uid 399); 19 Jul 2007 14:34:16 -0000 Received: from unknown (HELO DELL9400) (jack.toering@leadingedgeita.com@71.205.40.245) by mail2.hspheredns.com with ESMTP; 19 Jul 2007 14:34:16 -0000 X-Originating-IP: 71.205.40.245 From: "Jack Toering" To: References: <029401c7c8af$e5d19c60$6480010a@DELL9400> <20070718120302.GA11968@voi.aagh.net> <032101c7c93c$3fe17770$6480010a@DELL9400> <200707181344.32037.darcy@ok-connect.com> Date: Thu, 19 Jul 2007 10:34:17 -0400 Organization: Leading Edge IT Architects Message-ID: <007801c7ca11$e3175380$6480010a@DELL9400> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Mailer: Microsoft Office Outlook 11 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.3138 thread-index: AcfJfOK13a4srifyQyeIjST/CO7fzAAj3zaQ In-Reply-To: <200707181344.32037.darcy@ok-connect.com> X-Mailman-Approved-At: Thu, 19 Jul 2007 14:36:25 +0000 Subject: RE: Which SMP CPU for FreeBSD 6.2? 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 Jul 2007 14:34:23 -0000 >Another site you can look at for reviews which is far less game/windows/office oriented is tweakers.net< I found the one review in English. While the site is named tweakers.net, an English word, but most of the reviews and text seem to be in Friesian, Dutch, Flemish, or German or some dialect from the northern European mainland. I don't see a way to view the remainder of the content on the site in English unless I'm missing something. Thanks! -----Original Message----- From: Darcy Buskermolen [mailto:darcy@ok-connect.com] Sent: Wednesday, July 18, 2007 4:45 PM To: freebsd-performance@freebsd.org Cc: Jack Toering Subject: Re: Which SMP CPU for FreeBSD 6.2? On Wednesday 18 July 2007 06:05:00 Jack Toering wrote: > >You probably don't need to follow every move, but it might make sense > >to do > > some actual research if performance matters to you? You don't even > say what your performance critical server is going to be doing ;)< > > At the moment I'm not too sure either. :) The reason I say this is we > have a 246 Dual Opteron that is handling 15,000 unique visitors an > hour pulling data from around the world, generating weather maps, and pounding MySQL. > It has no problems. There is host a large hotel site on the same > server that is doing 900 and hour. 2/3 of the bandwidth is inbound to > make the calculations. It took us a lot off tweaking to get it where it is today. > > >but K8's are still pretty good with many workloads thanks to their > >far > > superior interconnects.< > > Which is why we went with dual 246s at the time. The second processor > on the Intel was a waste of money and it dual chip at the server level > couldn't hold a candle to their laptop chips and the cost more than AMD. > That decision was a no-brainer. > > >Most numbers I've seen have been more based on games and media > >encoding in > > single socket configurations, and in 32bit mode at that;< > > Yes, I'm quite skeptical about benchmarks even on web servers having > learned the hard way. We tried 3 big names on Linux where the site > fell down before 9AM. A 4th, Rackspace, the salesman didn't return my > calls after they had an engineer check out what we were doing. Before > and after we tried 3 versions off Linux ourselves with lots of > configurations. Based on all of the benchmarks, FreeBSD didn't make > sense which is why it took us so long to get here. Our competitors > are running 5 box server farms to do what we accomplish with this > setup and we have better response times. We are running SCSI raid. > There is no measurable user response difference between a load of 17 > or .7. Due to optimizations, our current load doesn't exceed 6 or 7 with normal being more around 2. > > Then the NOCs hated us because we used the bandwidth they promised and > we were told to move along one way or another 3 times. The best NOC > we've ever had is SAVVIS, Texas. They could not handle it either > where the server was at first, but after I proved to them with a > packet sniffer that there was a problem, they worked hard for us to > get us a place in the DC that could handle it. We've lived happily > ever after there. That's where the next one is going also. We've had > all we can handle of being a fugitive and a vagabond. > > Naturally, my thoughts are drifting toward, "Hey, the new site should > be doing the 246s and the old site something new." In the mean time > I've kept up off and on, on the issues with AMD code not being nearly > as efficient as the Intel code due to optimization for the Intel due > to its far greater numbers. Next, the abosolute 180 degree change in > the market between now and then concerning AMD and Intel. I've read > the Tom's hardware and AnandTech (I think) comparisions. However, a > lot of beautify theories get beat up by ugly realities and I have the > scars to prove it. That's why I'm here fishing. I'm not expecting to > find someone else with traffic and loads like this, but any insights > are welcome when it comes to rolling the dice again because anything > you can do to cut down the number of tries you need to throw the dice is a huge help. Another site you can look at for reviews which is far less game/windows/office oriented is tweakers.net, http://tweakers.net/reviews/661/7 for a good test of a woodcrest based server. > > Thanks! > > _______________________________________________ > 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" From owner-freebsd-performance@FreeBSD.ORG Thu Jul 19 19:25:02 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 55D1416A404 for ; Thu, 19 Jul 2007 19:25:02 +0000 (UTC) (envelope-from matthijs@groov.nl) Received: from smtp-vbr13.xs4all.nl (smtp-vbr13.xs4all.nl [194.109.24.33]) by mx1.freebsd.org (Postfix) with ESMTP id E252413C494 for ; Thu, 19 Jul 2007 19:25:01 +0000 (UTC) (envelope-from matthijs@groov.nl) Received: from deadlock (home.groov.nl [80.126.115.247]) by smtp-vbr13.xs4all.nl (8.13.8/8.13.8) with ESMTP id l6JJ9Ujf083612; Thu, 19 Jul 2007 21:09:31 +0200 (CEST) (envelope-from matthijs@groov.nl) Message-Id: <200707191909.l6JJ9Ujf083612@smtp-vbr13.xs4all.nl> From: "Matthijs Breemans" To: "'Jack Toering'" , Date: Thu, 19 Jul 2007 21:10:24 +0200 MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Mailer: Microsoft Office Outlook, Build 11.0.6353 In-Reply-To: <007801c7ca11$e3175380$6480010a@DELL9400> thread-index: AcfJfOK13a4srifyQyeIjST/CO7fzAAj3zaQAAr6RbA= X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.3138 X-Virus-Scanned: by XS4ALL Virus Scanner Cc: Subject: RE: Which SMP CPU for FreeBSD 6.2? 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 Jul 2007 19:25:02 -0000 Tweakers.net is a Dutch site, and you can translate it with Babblefish or some other translation tool. -----Oorspronkelijk bericht----- Van: owner-freebsd-performance@freebsd.org [mailto:owner-freebsd-performance@freebsd.org] Namens Jack Toering Verzonden: donderdag 19 juli 2007 16:34 Aan: freebsd-performance@freebsd.org Onderwerp: RE: Which SMP CPU for FreeBSD 6.2? >Another site you can look at for reviews which is far less game/windows/office oriented is tweakers.net< I found the one review in English. While the site is named tweakers.net, an English word, but most of the reviews and text seem to be in Friesian, Dutch, Flemish, or German or some dialect from the northern European mainland. I don't see a way to view the remainder of the content on the site in English unless I'm missing something. Thanks! -----Original Message----- From: Darcy Buskermolen [mailto:darcy@ok-connect.com] Sent: Wednesday, July 18, 2007 4:45 PM To: freebsd-performance@freebsd.org Cc: Jack Toering Subject: Re: Which SMP CPU for FreeBSD 6.2? On Wednesday 18 July 2007 06:05:00 Jack Toering wrote: > >You probably don't need to follow every move, but it might make sense > >to do > > some actual research if performance matters to you? You don't even > say what your performance critical server is going to be doing ;)< > > At the moment I'm not too sure either. :) The reason I say this is we > have a 246 Dual Opteron that is handling 15,000 unique visitors an > hour pulling data from around the world, generating weather maps, and pounding MySQL. > It has no problems. There is host a large hotel site on the same > server that is doing 900 and hour. 2/3 of the bandwidth is inbound to > make the calculations. It took us a lot off tweaking to get it where it is today. > > >but K8's are still pretty good with many workloads thanks to their > >far > > superior interconnects.< > > Which is why we went with dual 246s at the time. The second processor > on the Intel was a waste of money and it dual chip at the server level > couldn't hold a candle to their laptop chips and the cost more than AMD. > That decision was a no-brainer. > > >Most numbers I've seen have been more based on games and media > >encoding in > > single socket configurations, and in 32bit mode at that;< > > Yes, I'm quite skeptical about benchmarks even on web servers having > learned the hard way. We tried 3 big names on Linux where the site > fell down before 9AM. A 4th, Rackspace, the salesman didn't return my > calls after they had an engineer check out what we were doing. Before > and after we tried 3 versions off Linux ourselves with lots of > configurations. Based on all of the benchmarks, FreeBSD didn't make > sense which is why it took us so long to get here. Our competitors > are running 5 box server farms to do what we accomplish with this > setup and we have better response times. We are running SCSI raid. > There is no measurable user response difference between a load of 17 > or .7. Due to optimizations, our current load doesn't exceed 6 or 7 with normal being more around 2. > > Then the NOCs hated us because we used the bandwidth they promised and > we were told to move along one way or another 3 times. The best NOC > we've ever had is SAVVIS, Texas. They could not handle it either > where the server was at first, but after I proved to them with a > packet sniffer that there was a problem, they worked hard for us to > get us a place in the DC that could handle it. We've lived happily > ever after there. That's where the next one is going also. We've had > all we can handle of being a fugitive and a vagabond. > > Naturally, my thoughts are drifting toward, "Hey, the new site should > be doing the 246s and the old site something new." In the mean time > I've kept up off and on, on the issues with AMD code not being nearly > as efficient as the Intel code due to optimization for the Intel due > to its far greater numbers. Next, the abosolute 180 degree change in > the market between now and then concerning AMD and Intel. I've read > the Tom's hardware and AnandTech (I think) comparisions. However, a > lot of beautify theories get beat up by ugly realities and I have the > scars to prove it. That's why I'm here fishing. I'm not expecting to > find someone else with traffic and loads like this, but any insights > are welcome when it comes to rolling the dice again because anything > you can do to cut down the number of tries you need to throw the dice is a huge help. Another site you can look at for reviews which is far less game/windows/office oriented is tweakers.net, http://tweakers.net/reviews/661/7 for a good test of a woodcrest based server. > > Thanks! > > _______________________________________________ > 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" _______________________________________________ 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" From owner-freebsd-performance@FreeBSD.ORG Fri Jul 20 09:29:17 2007 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 8A93A16A419; Fri, 20 Jul 2007 09:29:17 +0000 (UTC) (envelope-from stas@FreeBSD.org) Received: from com1.ht-systems.ru (com1.ht-systems.ru [83.97.104.204]) by mx1.freebsd.org (Postfix) with ESMTP id 052E213C457; Fri, 20 Jul 2007 09:29:16 +0000 (UTC) (envelope-from stas@FreeBSD.org) Received: from [85.21.245.235] (helo=phonon.SpringDaemons.com) by com1.ht-systems.ru with esmtpa (Exim 4.62) (envelope-from ) id 1IBonS-0003lh-7n; Fri, 20 Jul 2007 13:29:15 +0400 Received: from localhost (localhost [127.0.0.1]) by phonon.SpringDaemons.com (Postfix) with SMTP id 02D7511404; Fri, 20 Jul 2007 13:28:40 +0400 (MSD) Date: Fri, 20 Jul 2007 13:28:31 +0400 From: Stanislav Sedov To: "Michael Vaughn" Message-Id: <20070720132831.87cd008f.stas@FreeBSD.org> In-Reply-To: <2f0146460707181512x3841af57l588e4d6e67bd5884@mail.gmail.com> References: <2f0146460707181512x3841af57l588e4d6e67bd5884@mail.gmail.com> Organization: The FreeBSD Project X-Mailer: carrier-pigeon X-Voice: +7 916 849 20 23 X-XMPP: ssedov@jabber.ru X-ICQ: 208105021 X-Yahoo: stanislav_sedov X-PGP-Fingerprint: F21E D6CC 5626 9609 6CE2 A385 2BF5 5993 EB26 9581 X-University: MEPhI Mime-Version: 1.0 Content-Type: multipart/signed; protocol="application/pgp-signature"; micalg="PGP-SHA1"; boundary="Signature=_Fri__20_Jul_2007_13_28_31_+0400_gtJnMl+38.=FuSxv" X-Spam-Flag: SKIP X-Spam-Yversion: Spamooborona 1.6.0 Cc: freebsd-hackers@freebsd.org, freebsd-performance@freebsd.org, freebsd-questions@freebsd.org Subject: Re: FreeBSD 6.2-STABLE && apache 2.2.4 = bad performance. Help! 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 Jul 2007 09:29:17 -0000 --Signature=_Fri__20_Jul_2007_13_28_31_+0400_gtJnMl+38.=FuSxv Content-Type: text/plain; charset=US-ASCII Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Wed, 18 Jul 2007 23:12:57 +0100 "Michael Vaughn" mentioned: > Hello everyone, >=20 > I am contacting -performance, -questions, and -hackers in the hope someone > helps me troubleshoot a problem with FreeBSD 6.2 and apache 2.2.4 >=20 Try to run truss(1) on any of apache processes and look what it's doing. --=20 Stanislav Sedov ST4096-RIPE --Signature=_Fri__20_Jul_2007_13_28_31_+0400_gtJnMl+38.=FuSxv Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.4 (FreeBSD) iD8DBQFGoIBHK/VZk+smlYERAqEHAJ9P0ig41tc5wcnZ9VR3wlpIHI+0HQCghDSh Q7On7AP5rtfF3D1yl+8Nvxk= =mxbR -----END PGP SIGNATURE----- --Signature=_Fri__20_Jul_2007_13_28_31_+0400_gtJnMl+38.=FuSxv--