From owner-freebsd-performance@FreeBSD.ORG Sun Oct 24 00:14:06 2010 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 B4EA0106566C for ; Sun, 24 Oct 2010 00:14:06 +0000 (UTC) (envelope-from david@catwhisker.org) Received: from albert.catwhisker.org (m209-73.dsl.rawbw.com [198.144.209.73]) by mx1.freebsd.org (Postfix) with ESMTP id 8527B8FC0A for ; Sun, 24 Oct 2010 00:14:06 +0000 (UTC) Received: from albert.catwhisker.org (localhost [127.0.0.1]) by albert.catwhisker.org (8.14.4/8.14.4) with ESMTP id o9O0E6Rn070553 for ; Sat, 23 Oct 2010 17:14:06 -0700 (PDT) (envelope-from david@albert.catwhisker.org) Received: (from david@localhost) by albert.catwhisker.org (8.14.4/8.14.4/Submit) id o9O0E51p070552 for freebsd-performance@freebsd.org; Sat, 23 Oct 2010 17:14:05 -0700 (PDT) (envelope-from david) Date: Sat, 23 Oct 2010 17:14:05 -0700 From: David Wolfskill To: freebsd-performance@freebsd.org Message-ID: <20101024001405.GT52404@albert.catwhisker.org> References: <20101020174854.GZ21226@albert.catwhisker.org> <4CBF8032.8000609@freebsd.org> <20101022231730.GP52404@albert.catwhisker.org> <20101022234817.GQ52404@albert.catwhisker.org> <4CC22758.3040201@freebsd.org> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="6axCafNXXMM8qu6Q" Content-Disposition: inline In-Reply-To: <4CC22758.3040201@freebsd.org> User-Agent: Mutt/1.4.2.3i Subject: Re: Possible evidence of performance regression for 8.1-S (vs. 7.1) 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, 24 Oct 2010 00:14:06 -0000 --6axCafNXXMM8qu6Q Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Fri, Oct 22, 2010 at 05:07:52PM -0700, Julian Elischer wrote: > On 10/22/10 4:48 PM, David Wolfskill wrote: > >On Fri, Oct 22, 2010 at 04:17:30PM -0700, David Wolfskill wrote: > >>On Wed, Oct 20, 2010 at 04:50:10PM -0700, Julian Elischer wrote: > >>>... > >>>try the 7.x machine but running the 8.x kernel.. i.e. change nothing, > >>>but boot the new kernel. > >>I just started that test, to run over the weekend. > >>... > >Moot point: the build fails. > > > that shouldn't happen > maybe you need to set the uname vars so that the build THINKS it is on 7.x OK; poking through the source tree, I found quite a few references to uname -- and several to kern.osreldate. I tried setting UNAME_r to a value that gegins with "7.1-RELEASE", and the build ran, completing with a status code of 0. I'm not especially confident that the result of the build was equivalent -- some of my statistics-gathering was a bit warped by the experience, but appears to be salvagable. I've fired off a sequence of 5 iterations of the test build, with the UNAME_r set as above; we should have answers (to the question of the rusage data for 8.1-S kernel and 7.1-R+ userland) by Monday. Thanks! > ... > it's not unusual to have to do this when cross building with a=20 > different kernel. :-} OK; I confess that it hadn 't occurred to me that this would be useful, let alone expected to work. (Sure, I expect some basic stuff to function in that environment, but with more "moving parts," I get a tad nervous.) > Running an 8.x kernel with the 7.x userland is a crucial part of=20 > debugging this. > we really can't do much more until we have the results. We should have them, as above. > If the build fails. it's either becasue the build detects the kernel=20 > rev (which we want to > stop with this) or because of a bug in 8.1. It's supposed to be=20 > downwards compatible. Well, I know that there were some changes made to our build in order to get it to even work under 8.x at all; I suppose I can be thankful that those seem to be based on uname values, vs. kern.osreldate. :-} Peace, david --=20 David H. Wolfskill david@catwhisker.org Depriving a girl or boy of an opportunity for education is evil. See http://www.catwhisker.org/~david/publickey.gpg for my public key. --6axCafNXXMM8qu6Q Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.16 (FreeBSD) iEYEARECAAYFAkzDekwACgkQmprOCmdXAD2b3ACggBdUAkz68/98S0PHVsraKJ5t t2IAoIJkKNbQBUSiZhTASusMIfMugA9o =aBfP -----END PGP SIGNATURE----- --6axCafNXXMM8qu6Q-- From owner-freebsd-performance@FreeBSD.ORG Mon Oct 25 16:55:50 2010 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 9B9C0106566B; Mon, 25 Oct 2010 16:55:50 +0000 (UTC) (envelope-from david@catwhisker.org) Received: from albert.catwhisker.org (m209-73.dsl.rawbw.com [198.144.209.73]) by mx1.freebsd.org (Postfix) with ESMTP id 688048FC12; Mon, 25 Oct 2010 16:55:48 +0000 (UTC) Received: from albert.catwhisker.org (localhost [127.0.0.1]) by albert.catwhisker.org (8.14.4/8.14.4) with ESMTP id o9PGtmE3011903; Mon, 25 Oct 2010 09:55:48 -0700 (PDT) (envelope-from david@albert.catwhisker.org) Received: (from david@localhost) by albert.catwhisker.org (8.14.4/8.14.4/Submit) id o9PGtmab011902; Mon, 25 Oct 2010 09:55:48 -0700 (PDT) (envelope-from david) Date: Mon, 25 Oct 2010 09:55:48 -0700 From: David Wolfskill To: Julian Elischer Message-ID: <20101025165548.GE1519@albert.catwhisker.org> References: <20101020174854.GZ21226@albert.catwhisker.org> <4CBF8032.8000609@freebsd.org> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="mR8QP4gmHujQHb1c" Content-Disposition: inline In-Reply-To: <4CBF8032.8000609@freebsd.org> User-Agent: Mutt/1.4.2.3i Cc: freebsd-performance@freebsd.org Subject: Re: Possible evidence of performance regression for 8.1-S (vs. 7.1) 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, 25 Oct 2010 16:55:50 -0000 --mR8QP4gmHujQHb1c Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Wed, Oct 20, 2010 at 04:50:10PM -0700, Julian Elischer wrote: > ... > try the 7.x machine but running the 8.x kernel.. i.e. change nothing,=20 > but boot the new kernel. > ... OK; here are results of previous tests, along with the above. As noted earlier, I needed to set the UNAME_r environmant variable in order for the build to succeed with the 7.x userland & 8.x kernel: start stop real user sys host os 1287436357 1287461948 25590.99 81502.22 18115.07 ref-8x 8.1-S 1287462797 1287488766 25969.26 81452.14 17920.14 ref-8x 8.1-S 1287489641 1287515287 25645.84 81548.40 18256.52 ref-8x 8.1-S 1287516151 1287541481 25329.64 81546.23 18294.10 ref-8x 8.1-S 1287542355 1287568599 26244.59 81431.47 17902.39 ref-8x 8.1-S 1287525363 1287546846 21483.13 82628.20 21703.09 ref-7x 7.1-R+ 1287548005 1287569100 21094.63 82853.19 22185.02 ref-7x 7.1-R+ 1287570300 1287591371 21071.33 82756.81 21943.22 ref-7x 7.1-R+ 1287592592 1287614103 21511.23 82637.30 21849.90 ref-7x 7.1-R+ 1287615323 1287636770 21446.42 82715.81 21708.97 ref-7x 7.1-R+ 1287710312 1287732046 21733.20 82688.01 22108.95 ref-8x 7.1-R+ 1287733360 1287754549 21188.88 82869.09 21890.83 ref-8x 7.1-R+ 1287755881 1287777566 21684.09 82772.50 21933.74 ref-8x 7.1-R+ 1287879508 1287905173 25665.03 81696.22 18278.42 ref-8x 8.1-S[7.1-R+ = user] 1287906039 1287931709 25669.48 81735.00 18265.78 ref-8x 8.1-S[7.1-R+ = user] 1287932573 1287958275 25700.99 81700.40 18246.04 ref-8x 8.1-S[7.1-R+ = user] 1287959144 1287984859 25714.74 81800.40 18346.57 ref-8x 8.1-S[7.1-R+ = user] 1287985748 1288011752 26004.33 81569.28 17967.68 ref-8x 8.1-S[7.1-R+ = user] The stanzas depict resource usage during each iteration of the workload under test (building software) under various conditions: * First, building on the ref-8x machine, running FreeBSD 8.1-STABLE r214029. * For comparison, building on the ref-7.x machine, running a lightly- patched 7.1-R. * As a reality check, building on the ref-8.x machine, running the above-cited 7.1-R (+patches) -- just to verify that there wasn't something obviously different in the hardware configurations or connections. * Finally, the requested 8.1-S kernel with 7.1-R+ userland (though I ran it on the ref-8x machine, as I had some other things to do with the ref-7.x machine). It appears to me that the last test runs show results that are just about identical to the "native" 8.1-S kernel+userland, so if I understand the logic correctly, that appears to implicate something in the 8.1-S kernel (or the default configuration for same). Peace, david --=20 David H. Wolfskill david@catwhisker.org Depriving a girl or boy of an opportunity for education is evil. See http://www.catwhisker.org/~david/publickey.gpg for my public key. --mR8QP4gmHujQHb1c Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.16 (FreeBSD) iEYEARECAAYFAkzFtpMACgkQmprOCmdXAD1UqACfWm+Wk1vEydHUXxePLtgaSaxo DakAn3ulpJ/i80IcVr2SoBvh1ryKFee7 =IcZZ -----END PGP SIGNATURE----- --mR8QP4gmHujQHb1c-- From owner-freebsd-performance@FreeBSD.ORG Tue Oct 26 05:09:36 2010 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 7AFF0106566C for ; Tue, 26 Oct 2010 05:09:36 +0000 (UTC) (envelope-from julian@freebsd.org) Received: from out-0.mx.aerioconnect.net (outd.internet-mail-service.net [216.240.47.227]) by mx1.freebsd.org (Postfix) with ESMTP id 5BBCF8FC0A for ; Tue, 26 Oct 2010 05:09:36 +0000 (UTC) Received: from idiom.com (postfix@mx0.idiom.com [216.240.32.160]) by out-0.mx.aerioconnect.net (8.13.8/8.13.8) with ESMTP id o9Q58rEV022085; Mon, 25 Oct 2010 22:08:54 -0700 X-Client-Authorized: MaGic Cook1e X-Client-Authorized: MaGic Cook1e Received: from julian-mac.elischer.org (72-254-86-56.client.stsn.net [72.254.86.56]) by idiom.com (Postfix) with ESMTP id CE7A02D6014; Mon, 25 Oct 2010 22:08:52 -0700 (PDT) Message-ID: <4CC662A1.9030708@freebsd.org> Date: Mon, 25 Oct 2010 22:09:53 -0700 From: Julian Elischer User-Agent: Mozilla/5.0 (Macintosh; U; PPC Mac OS X 10.4; en-US; rv:1.9.2.11) Gecko/20101013 Thunderbird/3.1.5 MIME-Version: 1.0 To: David Wolfskill References: <20101020174854.GZ21226@albert.catwhisker.org> <4CBF8032.8000609@freebsd.org> <20101025165548.GE1519@albert.catwhisker.org> In-Reply-To: <20101025165548.GE1519@albert.catwhisker.org> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Scanned-By: MIMEDefang 2.67 on 216.240.47.51 Cc: freebsd-performance@freebsd.org Subject: Re: Possible evidence of performance regression for 8.1-S (vs. 7.1) 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, 26 Oct 2010 05:09:36 -0000 On 10/25/10 9:55 AM, David Wolfskill wrote: > On Wed, Oct 20, 2010 at 04:50:10PM -0700, Julian Elischer wrote: >> ... >> try the 7.x machine but running the 8.x kernel.. i.e. change nothing, >> but boot the new kernel. >> ... > OK; here are results of previous tests, along with the above. As noted > earlier, I needed to set the UNAME_r environmant variable in order for > the build to succeed with the 7.x userland& 8.x kernel: > > start stop real user sys host os > 1287436357 1287461948 25590.99 81502.22 18115.07 ref-8x 8.1-S > 1287462797 1287488766 25969.26 81452.14 17920.14 ref-8x 8.1-S > 1287489641 1287515287 25645.84 81548.40 18256.52 ref-8x 8.1-S > 1287516151 1287541481 25329.64 81546.23 18294.10 ref-8x 8.1-S > 1287542355 1287568599 26244.59 81431.47 17902.39 ref-8x 8.1-S > > 1287525363 1287546846 21483.13 82628.20 21703.09 ref-7x 7.1-R+ > 1287548005 1287569100 21094.63 82853.19 22185.02 ref-7x 7.1-R+ > 1287570300 1287591371 21071.33 82756.81 21943.22 ref-7x 7.1-R+ > 1287592592 1287614103 21511.23 82637.30 21849.90 ref-7x 7.1-R+ > 1287615323 1287636770 21446.42 82715.81 21708.97 ref-7x 7.1-R+ > > 1287710312 1287732046 21733.20 82688.01 22108.95 ref-8x 7.1-R+ > 1287733360 1287754549 21188.88 82869.09 21890.83 ref-8x 7.1-R+ > 1287755881 1287777566 21684.09 82772.50 21933.74 ref-8x 7.1-R+ > > 1287879508 1287905173 25665.03 81696.22 18278.42 ref-8x 8.1-S[7.1-R+ user] > 1287906039 1287931709 25669.48 81735.00 18265.78 ref-8x 8.1-S[7.1-R+ user] > 1287932573 1287958275 25700.99 81700.40 18246.04 ref-8x 8.1-S[7.1-R+ user] > 1287959144 1287984859 25714.74 81800.40 18346.57 ref-8x 8.1-S[7.1-R+ user] > 1287985748 1288011752 26004.33 81569.28 17967.68 ref-8x 8.1-S[7.1-R+ user] > > > The stanzas depict resource usage during each iteration of the workload > under test (building software) under various conditions: > > * First, building on the ref-8x machine, running FreeBSD 8.1-STABLE r214029. > > * For comparison, building on the ref-7.x machine, running a lightly- > patched 7.1-R. > > * As a reality check, building on the ref-8.x machine, running the > above-cited 7.1-R (+patches) -- just to verify that there wasn't > something obviously different in the hardware configurations or > connections. > > * Finally, the requested 8.1-S kernel with 7.1-R+ userland (though I ran > it on the ref-8x machine, as I had some other things to do with the > ref-7.x machine). > > It appears to me that the last test runs show results that are just > about identical to the "native" 8.1-S kernel+userland, so if I > understand the logic correctly, that appears to implicate something in > the 8.1-S kernel (or the default configuration for same). yes, exactly. however the interesting thing is that while it took more wall-clock time, it took less system and user time. you might try the 4bsd scheduler to see what that does.. also, compare the configs of the two kernels > Peace, > david From owner-freebsd-performance@FreeBSD.ORG Tue Oct 26 11:29:17 2010 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 CF013106566B for ; Tue, 26 Oct 2010 11:29:17 +0000 (UTC) (envelope-from david@catwhisker.org) Received: from albert.catwhisker.org (m209-73.dsl.rawbw.com [198.144.209.73]) by mx1.freebsd.org (Postfix) with ESMTP id 89BD98FC18 for ; Tue, 26 Oct 2010 11:29:17 +0000 (UTC) Received: from albert.catwhisker.org (localhost [127.0.0.1]) by albert.catwhisker.org (8.14.4/8.14.4) with ESMTP id o9QBTHNH005179 for ; Tue, 26 Oct 2010 04:29:17 -0700 (PDT) (envelope-from david@albert.catwhisker.org) Received: (from david@localhost) by albert.catwhisker.org (8.14.4/8.14.4/Submit) id o9QBTHtc005178 for freebsd-performance@freebsd.org; Tue, 26 Oct 2010 04:29:17 -0700 (PDT) (envelope-from david) Date: Tue, 26 Oct 2010 04:29:16 -0700 From: David Wolfskill To: freebsd-performance@freebsd.org Message-ID: <20101026112916.GB2262@albert.catwhisker.org> References: <20101020174854.GZ21226@albert.catwhisker.org> <4CBF8032.8000609@freebsd.org> <20101025165548.GE1519@albert.catwhisker.org> <4CC662A1.9030708@freebsd.org> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="H1spWtNR+x+ondvy" Content-Disposition: inline In-Reply-To: <4CC662A1.9030708@freebsd.org> User-Agent: Mutt/1.4.2.3i Subject: Re: Possible evidence of performance regression for 8.1-S (vs. 7.1) 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, 26 Oct 2010 11:29:17 -0000 --H1spWtNR+x+ondvy Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Mon, Oct 25, 2010 at 10:09:53PM -0700, Julian Elischer wrote: > ... > >It appears to me that the last test runs show results that are just > >about identical to the "native" 8.1-S kernel+userland, so if I > >understand the logic correctly, that appears to implicate something in > >the 8.1-S kernel (or the default configuration for same). >=20 > yes, exactly. Good to get confirmation; thanks. :-} > however the interesting thing is that while it took more wall-clock time, > it took less system and user time. Aye; Dan Nelson also pointed that out, and it is rather interesting. > you might try the 4bsd scheduler to see what that does.. OK -- but we were using the default scheduler in each case. The basic point I'm making here is the apparent performance regression for similarly-configured systems under 7.1 vs. 8.1. > also, compare the configs of the two kernels Well, under 7.1, we used the MAC kernel config; since that didn't exist for 8.x, I used GENERIC for it. (We had used GENERIC under 7.1 until a certain application we use required MAC support. I haven't tried to make that application work under 8.x yet, as there are plans to deprecate its use.) Peace, david --=20 David H. Wolfskill david@catwhisker.org Depriving a girl or boy of an opportunity for education is evil. See http://www.catwhisker.org/~david/publickey.gpg for my public key. --H1spWtNR+x+ondvy Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.16 (FreeBSD) iEYEARECAAYFAkzGu4sACgkQmprOCmdXAD1fXgCfTHLzXJMJOOMwZvHzf118fbWp IyQAniUdlrey0PrmwkFDmPgzlmUkugtc =aVNY -----END PGP SIGNATURE----- --H1spWtNR+x+ondvy-- From owner-freebsd-performance@FreeBSD.ORG Tue Oct 26 14:34:12 2010 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 EF22E106566C for ; Tue, 26 Oct 2010 14:34:12 +0000 (UTC) (envelope-from mike@sentex.net) Received: from smarthost2.sentex.ca (smarthost2-6.sentex.ca [IPv6:2607:f3e0:80:80::2]) by mx1.freebsd.org (Postfix) with ESMTP id 784E88FC13 for ; Tue, 26 Oct 2010 14:34:12 +0000 (UTC) Received: from lava.sentex.ca (pyroxene.sentex.ca [199.212.134.18]) by smarthost2.sentex.ca (8.14.4/8.14.4) with ESMTP id o9QEY4Tf066332 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO) for ; Tue, 26 Oct 2010 10:34:04 -0400 (EDT) (envelope-from mike@sentex.net) Received: from mdt-xp.sentex.net (simeon.sentex.ca [192.168.43.27]) by lava.sentex.ca (8.14.4/8.14.4) with ESMTP id o9QEY32Q037618; Tue, 26 Oct 2010 10:34:03 -0400 (EDT) (envelope-from mike@sentex.net) Message-Id: <201010261434.o9QEY32Q037618@lava.sentex.ca> X-Mailer: QUALCOMM Windows Eudora Version 7.1.0.9 Date: Tue, 26 Oct 2010 10:34:08 -0400 To: David Wolfskill , freebsd-performance@freebsd.org From: Mike Tancsa In-Reply-To: <20101026112916.GB2262@albert.catwhisker.org> References: <20101020174854.GZ21226@albert.catwhisker.org> <4CBF8032.8000609@freebsd.org> <20101025165548.GE1519@albert.catwhisker.org> <4CC662A1.9030708@freebsd.org> <20101026112916.GB2262@albert.catwhisker.org> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed X-Scanned-By: MIMEDefang 2.67 on 205.211.164.50 Cc: Subject: Re: Possible evidence of performance regression for 8.1-S (vs. 7.1) 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, 26 Oct 2010 14:34:13 -0000 At 07:29 AM 10/26/2010, David Wolfskill wrote: >OK -- but we were using the default scheduler in each case. The basic >point I'm making here is the apparent performance regression for >similarly-configured systems under 7.1 vs. 8.1. ULE is the default in 7 as well. Perhaps remove some of the kernel options not in 7, that are in 8 by default? What is the disk subsystem ? just ata ? They seem innocuous enough, but worth a try options HWPMC_HOOKS # Necessary kernel hooks for hwpmc(4) options MAC # TrustedBSD MAC Framework options FLOWTABLE # per-cpu routing cache ---Mike -------------------------------------------------------------------- Mike Tancsa, tel +1 519 651 3400 Sentex Communications, mike@sentex.net Providing Internet since 1994 www.sentex.net Cambridge, Ontario Canada www.sentex.net/mike From owner-freebsd-performance@FreeBSD.ORG Tue Oct 26 15:30:12 2010 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 B7E20106573A for ; Tue, 26 Oct 2010 15:30:12 +0000 (UTC) (envelope-from david@catwhisker.org) Received: from albert.catwhisker.org (m209-73.dsl.rawbw.com [198.144.209.73]) by mx1.freebsd.org (Postfix) with ESMTP id 898BE8FC1F for ; Tue, 26 Oct 2010 15:30:11 +0000 (UTC) Received: from albert.catwhisker.org (localhost [127.0.0.1]) by albert.catwhisker.org (8.14.4/8.14.4) with ESMTP id o9QFUBMI007420 for ; Tue, 26 Oct 2010 08:30:11 -0700 (PDT) (envelope-from david@albert.catwhisker.org) Received: (from david@localhost) by albert.catwhisker.org (8.14.4/8.14.4/Submit) id o9QFUB4n007419 for freebsd-performance@freebsd.org; Tue, 26 Oct 2010 08:30:11 -0700 (PDT) (envelope-from david) Date: Tue, 26 Oct 2010 08:30:11 -0700 From: David Wolfskill To: freebsd-performance@freebsd.org Message-ID: <20101026153011.GE2262@albert.catwhisker.org> References: <20101020174854.GZ21226@albert.catwhisker.org> <4CBF8032.8000609@freebsd.org> <20101025165548.GE1519@albert.catwhisker.org> <4CC662A1.9030708@freebsd.org> <20101026112916.GB2262@albert.catwhisker.org> <201010261434.o9QEY32Q037618@lava.sentex.ca> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="Oiv9uiLrevHtW1RS" Content-Disposition: inline In-Reply-To: <201010261434.o9QEY32Q037618@lava.sentex.ca> User-Agent: Mutt/1.4.2.3i Subject: Re: Possible evidence of performance regression for 8.1-S (vs. 7.1) 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, 26 Oct 2010 15:30:12 -0000 --Oiv9uiLrevHtW1RS Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Tue, Oct 26, 2010 at 10:34:08AM -0400, Mike Tancsa wrote: > At 07:29 AM 10/26/2010, David Wolfskill wrote: >=20 > >OK -- but we were using the default scheduler in each case. The basic > >point I'm making here is the apparent performance regression for > >similarly-configured systems under 7.1 vs. 8.1. >=20 >=20 > ULE is the default in 7 as well. That was my recollection, yes. > Perhaps remove some of the kernel options not in 7, that are in > 8 by default? I'll look at that while I run some NFS-specific tests per ivoras@'s suggestion. > What is the disk subsystem ? just ata ? Errr... no. :-} ciss0: port 0xd800-0xd8ff mem 0xfb800000-0xfbbfffff,0= xfb7f f000-0xfb7fffff irq 30 at device 0.0 on pci6 ciss0: PERFORMANT Transport ciss0: [ITHREAD] =2E.. da0 at ciss0 bus 0 scbus0 target 0 lun 0 da0: Fixed Direct Access SCSI-5 device=20 da0: 135.168MB/s transfers da0: Command Queueing enabled da0: 429215MB (879032432 512 byte sectors: 255H 32S/T 65535C) da1 at ciss0 bus 0 scbus0 target 1 lun 0 da1: Fixed Direct Access SCSI-5 device=20 da1: 135.168MB/s transfers da1: Command Queueing enabled da1: 1716860MB (3516129392 512 byte sectors: 255H 32S/T 65535C) da2 at ciss0 bus 0 scbus0 target 2 lun 0 da2: Fixed Direct Access SCSI-5 device=20 da2: 135.168MB/s transfers da2: Command Queueing enabled da2: 1716860MB (3516129392 512 byte sectors: 255H 32S/T 65535C) da3 at ciss0 bus 0 scbus0 target 3 lun 0 da3: Fixed Direct Access SCSI-5 device=20 da3: 135.168MB/s transfers da3: Command Queueing enabled da3: 858430MB (1758064752 512 byte sectors: 255H 32S/T 65535C) The logical drive where the activity is taking place is /dev/da1, which is a 4-spindle RAID 0 group of 15Krpm SAS drives. CPU is: CPU: Intel(R) Xeon(R) CPU E5540 @ 2.53GHz (2533.44-MHz 686-class= CPU) Origin =3D "GenuineIntel" Id =3D 0x106a5 Family =3D 6 Model =3D 1a St= epping =3D 5 Features=3D0xbfebfbff Features2=3D0x9ce3bd AMD Features=3D0x28100000 AMD Features2=3D0x1 TSC: P-state invariant =2E.. FreeBSD/SMP: Multiprocessor System Detected: 8 CPUs FreeBSD/SMP: 2 package(s) x 4 core(s) cpu0 (BSP): APIC ID: 0 cpu1 (AP): APIC ID: 2 cpu2 (AP): APIC ID: 4 cpu3 (AP): APIC ID: 6 cpu4 (AP): APIC ID: 16 cpu5 (AP): APIC ID: 18 cpu6 (AP): APIC ID: 20 cpu7 (AP): APIC ID: 22 > They seem innocuous enough, but worth a try >=20 > options HWPMC_HOOKS # Necessary kernel hooks for hwpmc= (4) > options MAC # TrustedBSD MAC Framework > options FLOWTABLE # per-cpu routing cache > ... Well, MAC should probably stay, as we use the MAC kernel config for 7.1. Also, to get the "patched" 7.1-R, the following steps suffice (in case anyone is interested in attempting to reproduce the results): * svn co release/7.1.0. * svn merge -c186860 stable/7 * svn merge -c190970 stable/7 * svn merge -c203072 head * svn merge -c209964 stable/7 Remember to use MAC as your kernel config. Peace, david --=20 David H. Wolfskill david@catwhisker.org Depriving a girl or boy of an opportunity for education is evil. See http://www.catwhisker.org/~david/publickey.gpg for my public key. --Oiv9uiLrevHtW1RS Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.16 (FreeBSD) iEYEARECAAYFAkzG9AIACgkQmprOCmdXAD004QCgiSw1ekh4XT0SUdeqlDNIxIye uXQAn2lr4LUb9VP4TuxqAnqfRSjQ1l85 =orz7 -----END PGP SIGNATURE----- --Oiv9uiLrevHtW1RS-- From owner-freebsd-performance@FreeBSD.ORG Tue Oct 26 17:45:02 2010 Return-Path: Delivered-To: performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id D1CDA106566B for ; Tue, 26 Oct 2010 17:45:02 +0000 (UTC) (envelope-from david@catwhisker.org) Received: from albert.catwhisker.org (m209-73.dsl.rawbw.com [198.144.209.73]) by mx1.freebsd.org (Postfix) with ESMTP id 9F4578FC20 for ; Tue, 26 Oct 2010 17:45:02 +0000 (UTC) Received: from albert.catwhisker.org (localhost [127.0.0.1]) by albert.catwhisker.org (8.14.4/8.14.4) with ESMTP id o9QHj2ZL008369 for ; Tue, 26 Oct 2010 10:45:02 -0700 (PDT) (envelope-from david@albert.catwhisker.org) Received: (from david@localhost) by albert.catwhisker.org (8.14.4/8.14.4/Submit) id o9QHj147008368 for performance@freebsd.org; Tue, 26 Oct 2010 10:45:02 -0700 (PDT) (envelope-from david) Date: Tue, 26 Oct 2010 10:45:01 -0700 From: David Wolfskill To: performance@freebsd.org Message-ID: <20101026174501.GH2262@albert.catwhisker.org> References: <20101020174854.GZ21226@albert.catwhisker.org> <20101021215330.GA86224@dan.emsphone.com> <20101021224237.GG52404@albert.catwhisker.org> <4CC6C396.1010905@freebsd.org> <20101026121352.GC2262@albert.catwhisker.org> <20101020174854.GZ21226@albert.catwhisker.org> <20101021215330.GA86224@dan.emsphone.com> <20101021224237.GG52404@albert.catwhisker.org> <4CC6C396.1010905@freebsd.org> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="TBNym+cBXeFsS4Vs" Content-Disposition: inline In-Reply-To: <4CC6C396.1010905@freebsd.org> User-Agent: Mutt/1.4.2.3i Cc: Subject: Re: Possible evidence of performance regression for 8.1-S (vs. 7.1) 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, 26 Oct 2010 17:45:02 -0000 --TBNym+cBXeFsS4Vs Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Tue, Oct 26, 2010 at 02:03:34PM +0200, Ivan Voras wrote: > ... > Since you now have the two kernels readily available, can you rule out > NFS by just repeating the step which involves it in both kernels and > compare the results? On Tue, Oct 26, 2010 at 02:47:11PM +0200, Ivan Voras wrote: > ... > Couldn't you simply run blogbench or just copy/untar a full /usr/ports > tree on NFS that while booting either of the kernels? OK; here are timing results from 5 iterations each of running "tar xf", reading aa gzipped tarball of /usr/ports (updated earlier this morning) from NFS & writing it to the same local file system that I used for the previously-cited builds. Each of these was run on the 8.x reference machine, and each terminated with a status code of 0: start stop real user sys os 1288111167 1288111298 131.14 12.77 17.88 7.1-R+ 1288111359 1288111488 128.40 12.50 19.40 7.1-R+ 1288111546 1288111678 131.85 12.80 19.24 7.1-R+ 1288111739 1288111862 122.84 12.61 19.60 7.1-R+ 1288111919 1288112050 131.35 12.12 19.73 7.1-R+ 1288109542 1288109653 111.26 12.03 14.88 8.1-S [7.1-R+ userland] 1288109686 1288109783 97.38 12.83 13.28 8.1-S [7.1-R+ userland] 1288109811 1288109923 112.18 12.32 14.21 8.1-S [7.1-R+ userland] 1288109958 1288110063 104.94 12.21 14.34 8.1-S [7.1-R+ userland] 1288110094 1288110205 110.98 12.02 14.72 8.1-S [7.1-R+ userland] 1288112673 1288112768 94.88 9.41 12.87 8.1-S 1288112793 1288112885 92.37 9.16 13.32 8.1-S 1288112909 1288113009 100.25 9.21 13.72 8.1-S 1288113035 1288113126 90.52 9.38 13.38 8.1-S 1288113149 1288113249 100.23 8.91 14.00 8.1-S And here are some ministat summaries of the above: dwolf-bsd(8.1-S)[12] ls real_k7_u7 real_k8_u8 sys_k8_u7 user_k7_u7 user_k8_u8 real_k8_u7 sys_k7_u7 sys_k8_u8 user_k8_u7 dwolf-bsd(8.1-S)[13] ministat -s real* x real_k7_u7 + real_k8_u7 * real_k8_u8 +--------------------------------------------------------------------------= ----+ | * = | |* * * + * + +++ x x = xx | | |______A___= M__|| | |___________A_____M____| = | | |______MA_______| = | +--------------------------------------------------------------------------= ----+ N Min Max Median Avg Stddev x 5 122.84 131.85 131.14 129.116 3.7570906 + 5 97.38 112.18 110.98 107.348 6.2650395 Difference at 95.0% confidence -21.768 +/- 7.5337 -16.8593% +/- 5.83483% (Student's t, pooled s =3D 5.16558) * 5 90.52 100.25 94.88 95.65 4.466671 Difference at 95.0% confidence -33.466 +/- 6.01922 -25.9193% +/- 4.66187% (Student's t, pooled s =3D 4.12716) dwolf-bsd(8.1-S)[14] ministat -s user* x user_k7_u7 + user_k8_u7 * user_k8_u8 +--------------------------------------------------------------------------= ----+ |* ** ** + x + + x x= x+| | |_____AM= ___|| | |____MA______|= | | |___A___| = | +--------------------------------------------------------------------------= ----+ N Min Max Median Avg Stddev x 5 12.12 12.8 12.61 12.56 0.27449954 + 5 12.02 12.83 12.21 12.282 0.33131556 No difference proven at 95.0% confidence * 5 8.91 9.41 9.21 9.214 0.20082331 Difference at 95.0% confidence -3.346 +/- 0.350755 -26.6401% +/- 2.79263% (Student's t, pooled s =3D 0.240499) dwolf-bsd(8.1-S)[15] ministat -s sys* x sys_k7_u7 + sys_k8_u7 * sys_k8_u8 +--------------------------------------------------------------------------= ----+ |* +** * * ++ + + x xx = xx | | |_______A_M__= ___|| | |_____AM_____| = | | |___A____| = | +--------------------------------------------------------------------------= ----+ N Min Max Median Avg Stddev x 5 17.88 19.73 19.4 19.17 0.74505033 + 5 13.28 14.88 14.34 14.286 0.62488399 Difference at 95.0% confidence -4.884 +/- 1.00282 -25.4773% +/- 5.2312% (Student's t, pooled s =3D 0.687597) * 5 12.87 14 13.38 13.458 0.42815885 Difference at 95.0% confidence -5.712 +/- 0.886188 -29.7966% +/- 4.62279% (Student's t, pooled s =3D 0.607627) dwolf-bsd(8.1-S)[16]=20 Peace, david --=20 David H. Wolfskill david@catwhisker.org Depriving a girl or boy of an opportunity for education is evil. See http://www.catwhisker.org/~david/publickey.gpg for my public key. --TBNym+cBXeFsS4Vs Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.16 (FreeBSD) iEYEARECAAYFAkzHE50ACgkQmprOCmdXAD1gYQCeJbA/ffngJib88bCQRKQzccJH U2wAnjw3fpUq0lcnSDKBZWU061btBgZq =DL4O -----END PGP SIGNATURE----- --TBNym+cBXeFsS4Vs-- From owner-freebsd-performance@FreeBSD.ORG Wed Oct 27 09:54:15 2010 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 A88A710656A3 for ; Wed, 27 Oct 2010 09:54:15 +0000 (UTC) (envelope-from gofp-freebsd-performance@m.gmane.org) Received: from lo.gmane.org (lo.gmane.org [80.91.229.12]) by mx1.freebsd.org (Postfix) with ESMTP id 2A7738FC49 for ; Wed, 27 Oct 2010 09:54:14 +0000 (UTC) Received: from list by lo.gmane.org with local (Exim 4.69) (envelope-from ) id 1PB2hw-0001Un-Ps for freebsd-performance@freebsd.org; Wed, 27 Oct 2010 11:54:12 +0200 Received: from lara.cc.fer.hr ([161.53.72.113]) by main.gmane.org with esmtp (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Wed, 27 Oct 2010 11:54:12 +0200 Received: from ivoras by lara.cc.fer.hr with local (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Wed, 27 Oct 2010 11:54:12 +0200 X-Injected-Via-Gmane: http://gmane.org/ To: freebsd-performance@freebsd.org From: Ivan Voras Date: Wed, 27 Oct 2010 11:54:07 +0200 Lines: 45 Message-ID: References: <20101020174854.GZ21226@albert.catwhisker.org> <20101021215330.GA86224@dan.emsphone.com> <20101021224237.GG52404@albert.catwhisker.org> <4CC6C396.1010905@freebsd.org> <20101026121352.GC2262@albert.catwhisker.org> <20101020174854.GZ21226@albert.catwhisker.org> <20101021215330.GA86224@dan.emsphone.com> <20101021224237.GG52404@albert.catwhisker.org> <4CC6C396.1010905@freebsd.org> <20101026174501.GH2262@albert.catwhisker.org> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit X-Complaints-To: usenet@dough.gmane.org X-Gmane-NNTP-Posting-Host: lara.cc.fer.hr User-Agent: Mozilla/5.0 (X11; U; FreeBSD amd64; en-US; rv:1.9.1.12) Gecko/20101018 Thunderbird/3.0.8 In-Reply-To: <20101026174501.GH2262@albert.catwhisker.org> X-Enigmail-Version: 1.0.1 Subject: Re: Possible evidence of performance regression for 8.1-S (vs. 7.1) 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, 27 Oct 2010 09:54:15 -0000 On 10/26/10 19:45, David Wolfskill wrote: > On Tue, Oct 26, 2010 at 02:03:34PM +0200, Ivan Voras wrote: >> ... >> Since you now have the two kernels readily available, can you rule out >> NFS by just repeating the step which involves it in both kernels and >> compare the results? > > On Tue, Oct 26, 2010 at 02:47:11PM +0200, Ivan Voras wrote: >> ... >> Couldn't you simply run blogbench or just copy/untar a full /usr/ports >> tree on NFS that while booting either of the kernels? > > OK; here are timing results from 5 iterations each of running "tar > xf", reading aa gzipped tarball of /usr/ports (updated earlier this > morning) from NFS & writing it to the same local file system that > I used for the previously-cited builds. Each of these was run on > the 8.x reference machine, and each terminated with a status code of 0: > > start stop real user sys os > 1288111167 1288111298 131.14 12.77 17.88 7.1-R+ > 1288109542 1288109653 111.26 12.03 14.88 8.1-S [7.1-R+ userland] > 1288112673 1288112768 94.88 9.41 12.87 8.1-S There's a slight problem here: 8.1 with 7.1 userland should, in this test, behave the same as 8.1 with 8.1 userland. There have been no breakthroughs in gzip decompression or compiler optimization between those two releases that would make 8.1 userland faster. The most probable cause for the difference is simply disk drive location - inner vs outer tracks (you can run diskinfo -vt on the drive - it's non-destructive). This may also be the cause for your originally noticed speed difference. You could try some IO tuning on the box with sysctls like: vfs.hirunningspace=8388608 vfs.lorunningspace=6291456 vfs.ufs.dirhash_maxmem=8388608 vfs.read_max=32 ... but if the problem is due to the disk track locations, it's not really a problem. From owner-freebsd-performance@FreeBSD.ORG Wed Oct 27 10:55:07 2010 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 126F9106564A for ; Wed, 27 Oct 2010 10:55:07 +0000 (UTC) (envelope-from david@catwhisker.org) Received: from albert.catwhisker.org (m209-73.dsl.rawbw.com [198.144.209.73]) by mx1.freebsd.org (Postfix) with ESMTP id CF9198FC13 for ; Wed, 27 Oct 2010 10:55:06 +0000 (UTC) Received: from albert.catwhisker.org (localhost [127.0.0.1]) by albert.catwhisker.org (8.14.4/8.14.4) with ESMTP id o9RAt6KE013433 for ; Wed, 27 Oct 2010 03:55:06 -0700 (PDT) (envelope-from david@albert.catwhisker.org) Received: (from david@localhost) by albert.catwhisker.org (8.14.4/8.14.4/Submit) id o9RAt6Le013432 for freebsd-performance@freebsd.org; Wed, 27 Oct 2010 03:55:06 -0700 (PDT) (envelope-from david) Date: Wed, 27 Oct 2010 03:55:06 -0700 From: David Wolfskill To: freebsd-performance@freebsd.org Message-ID: <20101027105506.GD9443@albert.catwhisker.org> References: <20101021224237.GG52404@albert.catwhisker.org> <4CC6C396.1010905@freebsd.org> <20101026121352.GC2262@albert.catwhisker.org> <20101020174854.GZ21226@albert.catwhisker.org> <20101021215330.GA86224@dan.emsphone.com> <20101021224237.GG52404@albert.catwhisker.org> <4CC6C396.1010905@freebsd.org> <20101026174501.GH2262@albert.catwhisker.org> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="yudcn1FV7Hsu/q59" Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.4.2.3i Subject: Re: Possible evidence of performance regression for 8.1-S (vs. 7.1) 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, 27 Oct 2010 10:55:07 -0000 --yudcn1FV7Hsu/q59 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Wed, Oct 27, 2010 at 11:54:07AM +0200, Ivan Voras wrote: > ... > > the 8.x reference machine, and each terminated with a status code of 0: > >=20 > > start stop real user sys os > > 1288111167 1288111298 131.14 12.77 17.88 7.1-R+ >=20 > > 1288109542 1288109653 111.26 12.03 14.88 8.1-S [7.1-R+ userland] >=20 > > 1288112673 1288112768 94.88 9.41 12.87 8.1-S >=20 > There's a slight problem here: 8.1 with 7.1 userland should, in this > test, behave the same as 8.1 with 8.1 userland. There have been no > breakthroughs in gzip decompression or compiler optimization between > those two releases that would make 8.1 userland faster. I'm reporting what I measured. That said, we do have some history of variations in network activity (outside of anything directly involving systems under test), so it's possible that since these tests were so brief, each occurred in its own idiosyncratic "network ecology", thus skewing the reported results. I could re-run the tests easily enough, and increase the number of iterations (perhaps by a factor of 5 or 10).... > The most probable cause for the difference is simply disk drive location > - inner vs outer tracks (you can run diskinfo -vt on the drive - it's > non-destructive). This may also be the cause for your originally noticed > speed difference. I was using the same (4-spindle RAID 0 group) destination file system on the same machine for every one of those tests. Against what might I compare? > You could try some IO tuning on the box with sysctls like: >=20 > vfs.hirunningspace=3D8388608 > vfs.lorunningspace=3D6291456 > vfs.ufs.dirhash_maxmem=3D8388608 > vfs.read_max=3D32 Hmmm.... I'll look into these. > ... but if the problem is due to the disk track locations, it's not > really a problem. Well, as noted, it was the same file system each time. Precisely which blocks were allocated is not, as far as I know, not somethiong over which I have much influence or any control. But going back to the original issue -- recall, the above was intended to test just the NFS component, which was expected to be somewhere between "small" and "negligible" with respect to the originally-reported apparent discrepancy -- I was seeing consistent results of a "precious workload" taking a little under 6 hours under 7.1-R+, and a bit over 7 hours under vanilla 8.1-S. That *is* a problem, as I cannot justify a migration to a branch of FreeBSD that imposes about a 23% penalty in elapsed time on this workload. I want folks at work to have more reason to want to use (newer branches of) FreeBSD, not less. Peace, david --=20 David H. Wolfskill david@catwhisker.org Depriving a girl or boy of an opportunity for education is evil. See http://www.catwhisker.org/~david/publickey.gpg for my public key. --yudcn1FV7Hsu/q59 Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.16 (FreeBSD) iEYEARECAAYFAkzIBQkACgkQmprOCmdXAD130QCgh+rmcnCDKAkJYDH4SY2BgRoQ SKQAn36CbkOLGKh4G0lZwU3zeGycbY36 =2vL6 -----END PGP SIGNATURE----- --yudcn1FV7Hsu/q59-- From owner-freebsd-performance@FreeBSD.ORG Wed Oct 27 11:06:27 2010 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 A6532106566B for ; Wed, 27 Oct 2010 11:06:27 +0000 (UTC) (envelope-from gofp-freebsd-performance@m.gmane.org) Received: from lo.gmane.org (lo.gmane.org [80.91.229.12]) by mx1.freebsd.org (Postfix) with ESMTP id 59ACB8FC08 for ; Wed, 27 Oct 2010 11:06:27 +0000 (UTC) Received: from list by lo.gmane.org with local (Exim 4.69) (envelope-from ) id 1PB3po-0002SP-Iw for freebsd-performance@freebsd.org; Wed, 27 Oct 2010 13:06:24 +0200 Received: from lara.cc.fer.hr ([161.53.72.113]) by main.gmane.org with esmtp (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Wed, 27 Oct 2010 13:06:24 +0200 Received: from ivoras by lara.cc.fer.hr with local (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Wed, 27 Oct 2010 13:06:24 +0200 X-Injected-Via-Gmane: http://gmane.org/ To: freebsd-performance@freebsd.org From: Ivan Voras Date: Wed, 27 Oct 2010 13:06:18 +0200 Lines: 22 Message-ID: References: <20101021224237.GG52404@albert.catwhisker.org> <4CC6C396.1010905@freebsd.org> <20101026121352.GC2262@albert.catwhisker.org> <20101020174854.GZ21226@albert.catwhisker.org> <20101021215330.GA86224@dan.emsphone.com> <20101021224237.GG52404@albert.catwhisker.org> <4CC6C396.1010905@freebsd.org> <20101026174501.GH2262@albert.catwhisker.org> <20101027105506.GD9443@albert.catwhisker.org> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit X-Complaints-To: usenet@dough.gmane.org X-Gmane-NNTP-Posting-Host: lara.cc.fer.hr User-Agent: Mozilla/5.0 (X11; U; FreeBSD amd64; en-US; rv:1.9.1.12) Gecko/20101018 Thunderbird/3.0.8 In-Reply-To: <20101027105506.GD9443@albert.catwhisker.org> X-Enigmail-Version: 1.0.1 Subject: Re: Possible evidence of performance regression for 8.1-S (vs. 7.1) 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, 27 Oct 2010 11:06:27 -0000 On 10/27/10 12:55, David Wolfskill wrote: > That *is* a problem, as I cannot justify a migration to a branch > of FreeBSD that imposes about a 23% penalty in elapsed time on this > workload. I want folks at work to have more reason to want to use > (newer branches of) FreeBSD, not less. That is understandable, I'm only saying that given the symptoms - some "waiting" that is not accounted in system or user time and with NFS mostly ruled out (maybe a better test would be to extract a local tarball *to* the NFS server), the next suspect is the disk system. In case you never tried "diskinfo -vt", its results look like this: outside: 102400 kbytes in 1.222998 sec = 83729 kbytes/sec middle: 102400 kbytes in 1.448580 sec = 70690 kbytes/sec inside: 102400 kbytes in 2.531694 sec = 40447 kbytes/sec note 2x drop in performance between outer and inner tracks. It's not certain, especially since your results are repeatable and you always use exactly the same file system, but it's possible. Have you tried the sysctls I've posted? From owner-freebsd-performance@FreeBSD.ORG Wed Oct 27 11:19:05 2010 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 5178E1065679 for ; Wed, 27 Oct 2010 11:19:05 +0000 (UTC) (envelope-from david@catwhisker.org) Received: from albert.catwhisker.org (m209-73.dsl.rawbw.com [198.144.209.73]) by mx1.freebsd.org (Postfix) with ESMTP id 204D18FC0C for ; Wed, 27 Oct 2010 11:19:05 +0000 (UTC) Received: from albert.catwhisker.org (localhost [127.0.0.1]) by albert.catwhisker.org (8.14.4/8.14.4) with ESMTP id o9RBJ4fC014207 for ; Wed, 27 Oct 2010 04:19:04 -0700 (PDT) (envelope-from david@albert.catwhisker.org) Received: (from david@localhost) by albert.catwhisker.org (8.14.4/8.14.4/Submit) id o9RBJ4a0014206 for freebsd-performance@freebsd.org; Wed, 27 Oct 2010 04:19:04 -0700 (PDT) (envelope-from david) Date: Wed, 27 Oct 2010 04:19:04 -0700 From: David Wolfskill To: freebsd-performance@freebsd.org Message-ID: <20101027111904.GF9443@albert.catwhisker.org> References: <20101026121352.GC2262@albert.catwhisker.org> <20101020174854.GZ21226@albert.catwhisker.org> <20101021215330.GA86224@dan.emsphone.com> <20101021224237.GG52404@albert.catwhisker.org> <4CC6C396.1010905@freebsd.org> <20101026174501.GH2262@albert.catwhisker.org> <20101027105506.GD9443@albert.catwhisker.org> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="kR3zbvD4cgoYnS/6" Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.4.2.3i Subject: Re: Possible evidence of performance regression for 8.1-S (vs. 7.1) 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, 27 Oct 2010 11:19:05 -0000 --kR3zbvD4cgoYnS/6 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Wed, Oct 27, 2010 at 01:06:18PM +0200, Ivan Voras wrote: > On 10/27/10 12:55, David Wolfskill wrote: >=20 > > That *is* a problem, as I cannot justify a migration to a branch > > of FreeBSD that imposes about a 23% penalty in elapsed time on this > > workload. I want folks at work to have more reason to want to use > > (newer branches of) FreeBSD, not less. >=20 > That is understandable, :-} > I'm only saying that given the symptoms - some > "waiting" that is not accounted in system or user time and with NFS > mostly ruled out (maybe a better test would be to extract a local > tarball *to* the NFS server), I could run that test, sure.... > the next suspect is the disk system. In > case you never tried "diskinfo -vt", its results look like this: >=20 > outside: 102400 kbytes in 1.222998 sec =3D 83729 kbytes/sec > middle: 102400 kbytes in 1.448580 sec =3D 70690 kbytes/sec > inside: 102400 kbytes in 2.531694 sec =3D 40447 kbytes/sec >=20 > note 2x drop in performance between outer and inner tracks. OK, but I'm not sure how that's likely to work for a multi-spindle RAID 0 group.... > It's not > certain, especially since your results are repeatable and you always use > exactly the same file system, For the NFS tests just done. For the "real" tests, the same file system was used when I ran the tests on the same machine; (obviously) different machines have different (local) file systems. (Still, similarly- configured 4-spindle RAID 0 groups were used for the target file systems on each machine.) > but it's possible. Have you tried the sysctls I've posted? Err... Not yet, no. I just got up & read your message. I leave for work in just over 3 hours. :-} [While it is possible to login from home, it's a liitle more awkward than (say) logging in to the FreeBSD cluster, and this is the time of day when I reboot my laptop a few times -- it's presently running 9.0-CURRENT #23 r214372 & building 9.0-CURRENT #23 r214413; it just finished building & smoke-testing FreeBSD 8.1-STABLE #14 r214413.] Peace, david --=20 David H. Wolfskill david@catwhisker.org Depriving a girl or boy of an opportunity for education is evil. See http://www.catwhisker.org/~david/publickey.gpg for my public key. --kR3zbvD4cgoYnS/6 Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.16 (FreeBSD) iEYEARECAAYFAkzICqgACgkQmprOCmdXAD0fXACfWVaJ/a66ZkX0dsQYpw38nqr4 nBAAnR59+aF3mBCGJ+Iz3S0mun42FFIn =U0fT -----END PGP SIGNATURE----- --kR3zbvD4cgoYnS/6-- From owner-freebsd-performance@FreeBSD.ORG Wed Oct 27 11:25:05 2010 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 944421065693 for ; Wed, 27 Oct 2010 11:25:05 +0000 (UTC) (envelope-from gofp-freebsd-performance@m.gmane.org) Received: from lo.gmane.org (lo.gmane.org [80.91.229.12]) by mx1.freebsd.org (Postfix) with ESMTP id 44DD38FC13 for ; Wed, 27 Oct 2010 11:25:05 +0000 (UTC) Received: from list by lo.gmane.org with local (Exim 4.69) (envelope-from ) id 1PB47q-0003DR-4R for freebsd-performance@freebsd.org; Wed, 27 Oct 2010 13:25:02 +0200 Received: from lara.cc.fer.hr ([161.53.72.113]) by main.gmane.org with esmtp (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Wed, 27 Oct 2010 13:25:02 +0200 Received: from ivoras by lara.cc.fer.hr with local (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Wed, 27 Oct 2010 13:25:02 +0200 X-Injected-Via-Gmane: http://gmane.org/ To: freebsd-performance@freebsd.org From: Ivan Voras Date: Wed, 27 Oct 2010 13:24:54 +0200 Lines: 13 Message-ID: References: <20101026121352.GC2262@albert.catwhisker.org> <20101020174854.GZ21226@albert.catwhisker.org> <20101021215330.GA86224@dan.emsphone.com> <20101021224237.GG52404@albert.catwhisker.org> <4CC6C396.1010905@freebsd.org> <20101026174501.GH2262@albert.catwhisker.org> <20101027105506.GD9443@albert.catwhisker.org> <20101027111904.GF9443@albert.catwhisker.org> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit X-Complaints-To: usenet@dough.gmane.org X-Gmane-NNTP-Posting-Host: lara.cc.fer.hr User-Agent: Mozilla/5.0 (X11; U; FreeBSD amd64; en-US; rv:1.9.1.12) Gecko/20101018 Thunderbird/3.0.8 In-Reply-To: <20101027111904.GF9443@albert.catwhisker.org> X-Enigmail-Version: 1.0.1 Subject: Re: Possible evidence of performance regression for 8.1-S (vs. 7.1) 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, 27 Oct 2010 11:25:05 -0000 On 10/27/10 13:19, David Wolfskill wrote: >> note 2x drop in performance between outer and inner tracks. > > OK, but I'm not sure how that's likely to work for a multi-spindle RAID > 0 group.... Unless the RAID controller is trying to be overly smart (i.e. plays with fire) by somehow alternating stripe positions on drives, exactly the same as with single drives - stripe 0 on drive one will be logically joined with stripe 0 on the other drive, etc. From owner-freebsd-performance@FreeBSD.ORG Wed Oct 27 15:46:38 2010 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 BC8C41065675; Wed, 27 Oct 2010 15:46:38 +0000 (UTC) (envelope-from mike@sentex.net) Received: from smarthost2.sentex.ca (smarthost2-6.sentex.ca [IPv6:2607:f3e0:80:80::2]) by mx1.freebsd.org (Postfix) with ESMTP id 5C4F18FC12; Wed, 27 Oct 2010 15:46:38 +0000 (UTC) Received: from lava.sentex.ca (pyroxene.sentex.ca [199.212.134.18]) by smarthost2.sentex.ca (8.14.4/8.14.4) with ESMTP id o9RFkSq0074453 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Wed, 27 Oct 2010 11:46:28 -0400 (EDT) (envelope-from mike@sentex.net) Received: from mdt-xp.sentex.net (simeon.sentex.ca [192.168.43.27]) by lava.sentex.ca (8.14.4/8.14.4) with ESMTP id o9RFkRQn046373; Wed, 27 Oct 2010 11:46:27 -0400 (EDT) (envelope-from mike@sentex.net) Message-Id: <201010271546.o9RFkRQn046373@lava.sentex.ca> X-Mailer: QUALCOMM Windows Eudora Version 7.1.0.9 Date: Wed, 27 Oct 2010 11:46:33 -0400 To: Ivan Voras , freebsd-performance@freebsd.org From: Mike Tancsa In-Reply-To: References: <20101026121352.GC2262@albert.catwhisker.org> <20101020174854.GZ21226@albert.catwhisker.org> <20101021215330.GA86224@dan.emsphone.com> <20101021224237.GG52404@albert.catwhisker.org> <4CC6C396.1010905@freebsd.org> <20101026174501.GH2262@albert.catwhisker.org> <20101027105506.GD9443@albert.catwhisker.org> <20101027111904.GF9443@albert.catwhisker.org> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed X-Scanned-By: MIMEDefang 2.67 on 205.211.164.50 Cc: Subject: Re: Possible evidence of performance regression for 8.1-S (vs. 7.1) 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, 27 Oct 2010 15:46:38 -0000 I did some very simple numbers to compare RELENG_7 and RELENG_8 i386. I netbooted an i5 (650 @ 3.20GHz) box with a single ata disk that I newfs'd so the track layout is consistent across boots. I did a couple of very simple userland tests to make sure things are sane and they appear to be, at least with the apps I used. The only small difference I found was using sysbench on file io with RELENG_7. If there are some more synthetic benchmarks you want me to run, let me know as its easy to boot back and forth between OSes. But cpu and memory driven tests all seem to be about the same and the disk io differences are pretty small. sysbench --num-threads=16 --test=fileio --file-total-size=3G --file-test-mode=rndrw RELENG_7, RELENG_8 ATA, RELENG_8 AHCI, RELENG_8 AHCI with suggested sysctl changes Test execution summary: total time: 25.9904s,28.3793s,28.3160s,28.5659s total number of events: 10000 total time taken by event execution: 12.7917,14.4525,7.9162,5.2872 per-request statistics: min: 0.01ms,0.01ms,0.01ms,0.01ms avg: 1.28ms,1.45ms,0.79ms,0.53ms max: 96.97ms,102.47ms,72.74ms,62.44ms approx. 95 percentile: 0.78ms,7.52ms,3.06ms,1.42ms Threads fairness: events (avg/stddev): 625.0000/80.21,625.0000/99.54,625.0000/79.20,625.0000/98.04 execution time (avg/stddev): 0.7995/0.10,0.9033/0.11,0.4948/0.10,0.3304/0.11 Operations performed: 6001 Read, 3999 Write, 12672 Other = 22672 Total Read 93.766Mb Written 62.484Mb Total transferred 156.25Mb (6.0118Mb/sec) 384.76 Requests/sec executed Operations performed: 5997 Read, 4003 Write, 12800 Other = 22800 Total Read 93.703Mb Written 62.547Mb Total transferred 156.25Mb (5.5058Mb/sec) 352.37 Requests/sec executed Operations performed: 5998 Read, 4002 Write, 12800 Other = 22800 Total Read 93.719Mb Written 62.531Mb Total transferred 156.25Mb (5.5181Mb/sec) 353.16 Requests/sec executed Operations performed: 5999 Read, 4001 Write, 12672 Other = 22672 Total Read 93.734Mb Written 62.516Mb Total transferred 156.25Mb (5.4698Mb/sec) 350.07 Requests/sec executed sysbench --test=cpu --cpu-max-prime=20000 --num-threads=4 run sysbench 0.4.12: multi-threaded system evaluation benchmark Running the test with following options: Number of threads: 4 Doing CPU performance benchmark Threads started! Done. Maximum prime number checked in CPU test: 20000 Test execution summary: total time: 12.8866s,13.2981s total number of events: 10000 total time taken by event execution: 51.5295,53.1766 per-request statistics: min: 4.89ms,4.25ms avg: 5.15ms,5.32ms max: 5.22ms,5.38ms approx. 95 percentile: 5.16ms,5.32ms Threads fairness: events (avg/stddev): 2500.0000/1.73,2500.0000/1.73 execution time (avg/stddev): 12.8824/0.00,13.2942/0.00 sysbench --test=memory --num-threads=8 --memory-total-size=200000000000 run sysbench 0.4.12: multi-threaded system evaluation benchmark Running the test with following options: Number of threads: 8 Doing memory operations speed test Memory block size: 1K Memory transfer size: 2318M Memory operations type: write Memory scope type: global Threads started! Done. Operations performed: 2374516 (574580.71 ops/sec) 2318.86 MB transferred (561.11 MB/sec) Test execution summary: total time: 4.1326s,4.0980s total number of events: 2374516,2374516 total time taken by event execution: 23.6287,22.3643 per-request statistics: min: 0.00ms,0.00ms avg: 0.01ms,0.01ms max: 210.01ms,180.95ms approx. 95 percentile: 0.00ms,0.00ms Threads fairness: events (avg/stddev): 296814.5000/21552.12,296814.5000/19049.55 execution time (avg/stddev): 2.9536/0.17,2.7955/0.13 achinetboot# uname -a FreeBSD achinetboot.sentex.ca 7.3-STABLE FreeBSD 7.3-STABLE #0: Wed Oct 27 08:53:43 EDT 2010 mdtancsa@ich10.sentex.ca:/usr/RELENG7/obj/usr/RELENG7/src/sys/GENERIC i386 achinetboot# achinetboot# time cp /FreeBSD-8.1-RELEASE-i386-livefs.iso /mnt/ 0.000u 0.663s 0:22.19 2.9% 24+1310k 9+1998io 0pf+0w achinetboot# achinetboot# time pbzip2 -v FreeBSD-8.1-RELEASE-i386-livefs.iso Parallel BZIP2 v1.1.1 - by: Jeff Gilchrist [http://compression.ca] [Apr. 17, 2010] (uses libbzip2 by Julian Seward) Major contributions: Yavor Nikolov # CPUs: 4 BWT Block Size: 900 KB File Block Size: 900 KB Maximum Memory: 100 MB ------------------------------------------- File #: 1 of 1 Input Name: FreeBSD-8.1-RELEASE-i386-livefs.iso Output Name: FreeBSD-8.1-RELEASE-i386-livefs.iso.bz2 Input Size: 261816320 bytes Compressing data... Output Size: 92907893 bytes ------------------------------------------- Wall Clock: 18.624642 seconds 70.096u 1.607s 0:18.70 383.3% 74+1076k 2+709io 75pf+0w achinetboot# achinetboot# umount /mnt achinetboot# mount /dev/ad12 /mnt achinetboot# cd /mnt achinetboot# ls -l total 90802 drwxrwxr-x 2 root operator 512 Oct 27 13:11 .snap -rw-r--r-- 1 root wheel 92907893 Oct 27 13:27 FreeBSD-8.1-RELEASE-i386-livefs.iso.bz2 achinetboot# time pbzip2 -d FreeBSD-8.1-RELEASE-i386-livefs.iso.bz2 24.244u 0.985s 0:06.85 368.1% 75+1083k 724+1998io 0pf+0w achinetboot# 0(achinetboot)# uname -a FreeBSD achinetboot.sentex.ca 8.1-STABLE FreeBSD 8.1-STABLE #1: Wed Oct 13 10:50:14 EDT 2010 mdtancsa@ich10.sentex.ca:/usr/obj/usr/src/sys/pxe i386 0(achinetboot)# 0(achinetboot)# mount /dev/ada0 /mnt 0(achinetboot)# time cp /FreeBSD-8.1-RELEASE-i386-livefs.iso /mnt 0.000u 0.365s 0:22.43 1.6% 16+898k 9+1998io 0pf+0w 0(achinetboot)# 0(achinetboot)# time pbzip2 -v FreeBSD-8.1-RELEASE-i386-livefs.iso Parallel BZIP2 v1.1.1 - by: Jeff Gilchrist [http://compression.ca] [Apr. 17, 2010] (uses libbzip2 by Julian Seward) Major contributions: Yavor Nikolov # CPUs: 4 BWT Block Size: 900 KB File Block Size: 900 KB Maximum Memory: 100 MB ------------------------------------------- File #: 1 of 1 Input Name: FreeBSD-8.1-RELEASE-i386-livefs.iso Output Name: FreeBSD-8.1-RELEASE-i386-livefs.iso.bz2 Input Size: 261816320 bytes Compressing data... Output Size: 92907893 bytes ------------------------------------------- Wall Clock: 17.711059 seconds 68.352u 1.442s 0:17.84 391.1% 74+1075k 2+709io 22pf+0w 0(achinetboot)# 0(achinetboot)# umount /mnt 0(achinetboot)# mount /dev/ada0 /mnt 0(achinetboot)# cd /mnt 0(achinetboot)# time pbzip2 -d FreeBSD-8.1-RELEASE-i386-livefs.iso.bz2 23.510u 1.037s 0:06.62 370.6% 74+1075k 724+1998io 0pf+0w 0(achinetboot)# and RELENG_8 using ata 0(achinetboot)# mount /dev/ad12 /mnt 0(achinetboot)# time cp /FreeBSD-8.1-RELEASE-i386-livefs.iso /mnt/ 0.000u 0.356s 0:22.43 1.5% 22+1166k 8+1998io 0pf+0w 0(achinetboot)# 0(achinetboot)# time pbzip2 -v FreeBSD-8.1-RELEASE-i386-livefs.iso Parallel BZIP2 v1.1.1 - by: Jeff Gilchrist [http://compression.ca] [Apr. 17, 2010] (uses libbzip2 by Julian Seward) Major contributions: Yavor Nikolov # CPUs: 4 BWT Block Size: 900 KB File Block Size: 900 KB Maximum Memory: 100 MB ------------------------------------------- File #: 1 of 1 Input Name: FreeBSD-8.1-RELEASE-i386-livefs.iso Output Name: FreeBSD-8.1-RELEASE-i386-livefs.iso.bz2 Input Size: 261816320 bytes Compressing data... Output Size: 92907893 bytes ------------------------------------------- Wall Clock: 17.813195 seconds 68.003u 1.405s 0:17.81 389.6% 74+1074k 0+709io 0pf+0w 0(achinetboot)# To compile perl 5.10 outside of the ports, RELENG8 91.94 real 79.85 user 5.75 sys RELENG7 96.41 real 80.78 user 5.67 sys # -------------------------------------------------------------------- Mike Tancsa, tel +1 519 651 3400 Sentex Communications, mike@sentex.net Providing Internet since 1994 www.sentex.net Cambridge, Ontario Canada www.sentex.net/mike From owner-freebsd-performance@FreeBSD.ORG Wed Oct 27 16:34:18 2010 Return-Path: Delivered-To: performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 87156106574B for ; Wed, 27 Oct 2010 16:34:18 +0000 (UTC) (envelope-from david@catwhisker.org) Received: from albert.catwhisker.org (m209-73.dsl.rawbw.com [198.144.209.73]) by mx1.freebsd.org (Postfix) with ESMTP id 596A08FC12 for ; Wed, 27 Oct 2010 16:34:18 +0000 (UTC) Received: from albert.catwhisker.org (localhost [127.0.0.1]) by albert.catwhisker.org (8.14.4/8.14.4) with ESMTP id o9RGYHwc016443 for ; Wed, 27 Oct 2010 09:34:17 -0700 (PDT) (envelope-from david@albert.catwhisker.org) Received: (from david@localhost) by albert.catwhisker.org (8.14.4/8.14.4/Submit) id o9RGYH63016442 for performance@freebsd.org; Wed, 27 Oct 2010 09:34:17 -0700 (PDT) (envelope-from david) Date: Wed, 27 Oct 2010 09:34:17 -0700 From: David Wolfskill To: performance@freebsd.org Message-ID: <20101027163417.GI9443@albert.catwhisker.org> References: <20101021215330.GA86224@dan.emsphone.com> <20101021224237.GG52404@albert.catwhisker.org> <4CC6C396.1010905@freebsd.org> <20101026174501.GH2262@albert.catwhisker.org> <20101027105506.GD9443@albert.catwhisker.org> <20101027111904.GF9443@albert.catwhisker.org> <201010271546.o9RFkRQn046373@lava.sentex.ca> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="RwxaKO075aXzzOz0" Content-Disposition: inline In-Reply-To: <201010271546.o9RFkRQn046373@lava.sentex.ca> User-Agent: Mutt/1.4.2.3i Cc: Subject: Re: Possible evidence of performance regression for 8.1-S (vs. 7.1) 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, 27 Oct 2010 16:34:18 -0000 --RwxaKO075aXzzOz0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Wed, Oct 27, 2010 at 11:46:33AM -0400, Mike Tancsa wrote: >=20 > I did some very simple numbers to compare RELENG_7 and RELENG_8=20 > i386.... But cpu and=20 > memory driven tests all seem to be about the same and the disk io=20 > differences are pretty small. > ... This is good to know, but please note that what I was testing and reporting was: * release/7.1.0, with the following merged in: r186860 from stable/7 r190970 from stable/7 r203072 from head r209964 from stable/7 and using the MAC kernel config * stable/8 @r214029 using the GENERIC kernel config Peace, david --=20 David H. Wolfskill david@catwhisker.org Depriving a girl or boy of an opportunity for education is evil. See http://www.catwhisker.org/~david/publickey.gpg for my public key. --RwxaKO075aXzzOz0 Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.16 (FreeBSD) iEYEARECAAYFAkzIVIkACgkQmprOCmdXAD1rAACcDWLy5svJfo4Vnex+HIdQtJqe occAn1DJSgadHSgj4//wKmhr1appmsRi =Fcnw -----END PGP SIGNATURE----- --RwxaKO075aXzzOz0-- From owner-freebsd-performance@FreeBSD.ORG Wed Oct 27 17:05:21 2010 Return-Path: Delivered-To: performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 0C25B106566C for ; Wed, 27 Oct 2010 17:05:21 +0000 (UTC) (envelope-from mike@sentex.net) Received: from smarthost2.sentex.ca (smarthost2-6.sentex.ca [IPv6:2607:f3e0:80:80::2]) by mx1.freebsd.org (Postfix) with ESMTP id 586558FC13 for ; Wed, 27 Oct 2010 17:05:20 +0000 (UTC) Received: from lava.sentex.ca (pyroxene.sentex.ca [199.212.134.18]) by smarthost2.sentex.ca (8.14.4/8.14.4) with ESMTP id o9RH5A20084633 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO) for ; Wed, 27 Oct 2010 13:05:11 -0400 (EDT) (envelope-from mike@sentex.net) Received: from mdt-xp.sentex.net (simeon.sentex.ca [192.168.43.27]) by lava.sentex.ca (8.14.4/8.14.4) with ESMTP id o9RH5Auc046763; Wed, 27 Oct 2010 13:05:10 -0400 (EDT) (envelope-from mike@sentex.net) Message-Id: <201010271705.o9RH5Auc046763@lava.sentex.ca> X-Mailer: QUALCOMM Windows Eudora Version 7.1.0.9 Date: Wed, 27 Oct 2010 13:05:16 -0400 To: David Wolfskill , performance@freebsd.org From: Mike Tancsa In-Reply-To: <20101027163417.GI9443@albert.catwhisker.org> References: <20101021215330.GA86224@dan.emsphone.com> <20101021224237.GG52404@albert.catwhisker.org> <4CC6C396.1010905@freebsd.org> <20101026174501.GH2262@albert.catwhisker.org> <20101027105506.GD9443@albert.catwhisker.org> <20101027111904.GF9443@albert.catwhisker.org> <201010271546.o9RFkRQn046373@lava.sentex.ca> <20101027163417.GI9443@albert.catwhisker.org> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed X-Scanned-By: MIMEDefang 2.67 on 205.211.164.50 Cc: Subject: Re: Possible evidence of performance regression for 8.1-S (vs. 7.1) 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, 27 Oct 2010 17:05:21 -0000 At 12:34 PM 10/27/2010, David Wolfskill wrote: >* release/7.1.0, with the following merged in: > r186860 from stable/7 > r190970 from stable/7 > r203072 from head > r209964 from stable/7 > and using the MAC kernel config > >* stable/8 @r214029 using the GENERIC kernel config OK. I was using GENERIC vs GENERIC from RELENG7 and RELENG8 from today. MAC is in by default in RELENG_8 ---Mike >Peace, >david >-- >David H. Wolfskill david@catwhisker.org >Depriving a girl or boy of an opportunity for education is evil. > >See http://www.catwhisker.org/~david/publickey.gpg for my public key. > -------------------------------------------------------------------- Mike Tancsa, tel +1 519 651 3400 Sentex Communications, mike@sentex.net Providing Internet since 1994 www.sentex.net Cambridge, Ontario Canada www.sentex.net/mike From owner-freebsd-performance@FreeBSD.ORG Wed Oct 27 19:36:46 2010 Return-Path: Delivered-To: performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 24099106566B; Wed, 27 Oct 2010 19:36:46 +0000 (UTC) (envelope-from mike@sentex.net) Received: from smarthost2.sentex.ca (smarthost2-6.sentex.ca [IPv6:2607:f3e0:80:80::2]) by mx1.freebsd.org (Postfix) with ESMTP id CDD3B8FC12; Wed, 27 Oct 2010 19:36:45 +0000 (UTC) Received: from lava.sentex.ca (pyroxene.sentex.ca [199.212.134.18]) by smarthost2.sentex.ca (8.14.4/8.14.4) with ESMTP id o9RJabLu099424 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Wed, 27 Oct 2010 15:36:37 -0400 (EDT) (envelope-from mike@sentex.net) Received: from mdt-xp.sentex.net (simeon.sentex.ca [192.168.43.27]) by lava.sentex.ca (8.14.4/8.14.4) with ESMTP id o9RJaaDF047543; Wed, 27 Oct 2010 15:36:36 -0400 (EDT) (envelope-from mike@sentex.net) Message-Id: <201010271936.o9RJaaDF047543@lava.sentex.ca> X-Mailer: QUALCOMM Windows Eudora Version 7.1.0.9 Date: Wed, 27 Oct 2010 15:36:42 -0400 To: Ivan Voras , performance@freebsd.org From: Mike Tancsa In-Reply-To: <201010271705.o9RH5Auc046763@lava.sentex.ca> References: <20101021215330.GA86224@dan.emsphone.com> <20101021224237.GG52404@albert.catwhisker.org> <4CC6C396.1010905@freebsd.org> <20101026174501.GH2262@albert.catwhisker.org> <20101027105506.GD9443@albert.catwhisker.org> <20101027111904.GF9443@albert.catwhisker.org> <201010271546.o9RFkRQn046373@lava.sentex.ca> <20101027163417.GI9443@albert.catwhisker.org> <201010271705.o9RH5Auc046763@lava.sentex.ca> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed X-Scanned-By: MIMEDefang 2.67 on 205.211.164.50 Cc: Subject: Re: Possible evidence of performance regression for 8.1-S (vs. 7.1) 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, 27 Oct 2010 19:36:46 -0000 At 01:05 PM 10/27/2010, Mike Tancsa wrote: >At 12:34 PM 10/27/2010, David Wolfskill wrote: > >>* release/7.1.0, with the following merged in: >> r186860 from stable/7 >> r190970 from stable/7 >> r203072 from head >> r209964 from stable/7 >> and using the MAC kernel config >> >>* stable/8 @r214029 using the GENERIC kernel config > >OK. >I was using GENERIC vs GENERIC from RELENG7 and RELENG8 from >today. MAC is in by default in RELENG_8 I ran the file io tests again 5 times RELENG7, Operations performed: 6000 Read, 4000 Write, 12672 Other = 22672 Total Read 93.75Mb Written 62.5Mb Total transferred 156.25Mb (6.1136Mb/sec) 391.27 Requests/sec executed Test execution summary: total time: 25.5578s total number of events: 10000 total time taken by event execution: 12.0346 per-request statistics: min: 0.01ms avg: 1.20ms max: 101.39ms approx. 95 percentile: 4.77ms Threads fairness: events (avg/stddev): 625.0000/97.89 execution time (avg/stddev): 0.7522/0.15 RELENG7 times for 5 runs 25.5578 26.5351 25.5734 27.0934 26.6094 RELENG8, ATA driver times for 5 runs 28.9372 27.3766 29.3507 28.3406 28.0904 RELENG7 throughput (MB/s) 6.1136 5.8884 6.1099 5.7671 5.8720 RELENG_8 ATA throughput (MB/s) 5.3996 5.7074 5.3236 5.5133 5.5624 ministat -s -w 60 7 8 x 7 + 8 +------------------------------------------------------------+ |+ + + + + x xx x| | |______M____A__________| | | |__________AM_________| | +------------------------------------------------------------+ N Min Max Median Avg Stddev x 5 5.7671 6.1136 5.8884 5.9502 0.15464778 + 5 5.3236 5.7074 5.5133 5.50126 0.1485552 Difference at 95.0% confidence -0.44894 +/- 0.221147 -7.54496% +/- 3.71663% (Student's t, pooled s = 0.151632) -------------------------------------------------------------------- Mike Tancsa, tel +1 519 651 3400 Sentex Communications, mike@sentex.net Providing Internet since 1994 www.sentex.net Cambridge, Ontario Canada www.sentex.net/mike From owner-freebsd-performance@FreeBSD.ORG Wed Oct 27 20:19:04 2010 Return-Path: Delivered-To: performance@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 336A4106564A; Wed, 27 Oct 2010 20:19:04 +0000 (UTC) (envelope-from mike@sentex.net) Received: from smarthost2.sentex.ca (smarthost2-6.sentex.ca [IPv6:2607:f3e0:80:80::2]) by mx1.freebsd.org (Postfix) with ESMTP id F2C468FC18; Wed, 27 Oct 2010 20:19:03 +0000 (UTC) Received: from lava.sentex.ca (pyroxene.sentex.ca [199.212.134.18]) by smarthost2.sentex.ca (8.14.4/8.14.4) with ESMTP id o9RKItO8004555 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Wed, 27 Oct 2010 16:18:55 -0400 (EDT) (envelope-from mike@sentex.net) Received: from mdt-xp.sentex.net (simeon.sentex.ca [192.168.43.27]) by lava.sentex.ca (8.14.4/8.14.4) with ESMTP id o9RKItXi047758; Wed, 27 Oct 2010 16:18:55 -0400 (EDT) (envelope-from mike@sentex.net) Message-Id: <201010272018.o9RKItXi047758@lava.sentex.ca> X-Mailer: QUALCOMM Windows Eudora Version 7.1.0.9 Date: Wed, 27 Oct 2010 16:19:01 -0400 To: Ivan Voras , performance@freebsd.org From: Mike Tancsa In-Reply-To: <201010271936.o9RJaaDF047543@lava.sentex.ca> References: <20101021215330.GA86224@dan.emsphone.com> <20101021224237.GG52404@albert.catwhisker.org> <4CC6C396.1010905@freebsd.org> <20101026174501.GH2262@albert.catwhisker.org> <20101027105506.GD9443@albert.catwhisker.org> <20101027111904.GF9443@albert.catwhisker.org> <201010271546.o9RFkRQn046373@lava.sentex.ca> <20101027163417.GI9443@albert.catwhisker.org> <201010271705.o9RH5Auc046763@lava.sentex.ca> <201010271936.o9RJaaDF047543@lava.sentex.ca> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed X-Scanned-By: MIMEDefang 2.67 on 205.211.164.50 Cc: Subject: Re: Possible evidence of performance regression for 8.1-S (vs. 7.1) 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, 27 Oct 2010 20:19:04 -0000 At 03:36 PM 10/27/2010, Mike Tancsa wrote: >At 01:05 PM 10/27/2010, Mike Tancsa wrote: >>At 12:34 PM 10/27/2010, David Wolfskill wrote: >> >>>* release/7.1.0, with the following merged in: >>> r186860 from stable/7 >>> r190970 from stable/7 >>> r203072 from head >>> r209964 from stable/7 >>> and using the MAC kernel config >>> >>>* stable/8 @r214029 using the GENERIC kernel config >> >>OK. >>I was using GENERIC vs GENERIC from RELENG7 and RELENG8 from >>today. MAC is in by default in RELENG_8 One bright spot on the file IO is the AHCI driver when the test is done with much larger data. Instead of 3G of test data I ran 6G.... I am guessing AHCI does better due to the command re-ordering RELENG8 Running the test with following options: Number of threads: 16 Extra file open flags: 0 128 files, 48Mb each 6Gb total file size Block size 16Kb Number of random requests for random IO: 10000 Read/Write ratio for combined random IO test: 1.50 Periodic FSYNC enabled, calling fsync() each 100 requests. Calling fsync() at the end of test, Enabled. Using synchronous I/O mode Doing random r/w test Threads started! Done. RELENG8 (AHCI) Operations performed: 6005 Read, 3995 Write, 12800 Other = 22800 Total Read 93.828Mb Written 62.422Mb Total transferred 156.25Mb (2.306Mb/sec) 147.58 Requests/sec executed Read 93.844Mb Written 62.406Mb Total transferred 156.25Mb (3.182Mb/sec) 203.68 Requests/sec executed Read 93.75Mb Written 62.5Mb Total transferred 156.25Mb (5.039Mb/sec) 322.53 Requests/sec executed Read 93.781Mb Written 62.469Mb Total transferred 156.25Mb (3.470Mb/sec) 222.12 Requests/sec executed Read 93.719Mb Written 62.531Mb Total transferred 156.25Mb (4.839Mb/sec) 309.72 Requests/sec executed vs RELENG7 Read 93.781Mb Written 62.469Mb Total transferred 156.25Mb (1.52Mb/sec) 97.67 Requests/sec executed Read 93.812Mb Written 62.438Mb Total transferred 156.25Mb (2.031Mb/sec) 130.00 Requests/sec executed Read 93.781Mb Written 62.469Mb Total transferred 156.25Mb (1.834Mb/sec) 117.40 Requests/sec executed Read 93.75Mb Written 62.5Mb Total transferred 156.25Mb (5.065Mb/sec) 324.16 Requests/sec executed Read 93.797Mb Written 62.453Mb Total transferred 156.25Mb (4.960Mb/sec) 317.46 Requests/sec executed ---Mike -------------------------------------------------------------------- Mike Tancsa, tel +1 519 651 3400 Sentex Communications, mike@sentex.net Providing Internet since 1994 www.sentex.net Cambridge, Ontario Canada www.sentex.net/mike From owner-freebsd-performance@FreeBSD.ORG Fri Oct 29 16:58:53 2010 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 88C91106564A for ; Fri, 29 Oct 2010 16:58:53 +0000 (UTC) (envelope-from david@catwhisker.org) Received: from albert.catwhisker.org (m209-73.dsl.rawbw.com [198.144.209.73]) by mx1.freebsd.org (Postfix) with ESMTP id 426E38FC1B for ; Fri, 29 Oct 2010 16:58:52 +0000 (UTC) Received: from albert.catwhisker.org (localhost [127.0.0.1]) by albert.catwhisker.org (8.14.4/8.14.4) with ESMTP id o9TGwqND017153 for ; Fri, 29 Oct 2010 09:58:52 -0700 (PDT) (envelope-from david@albert.catwhisker.org) Received: (from david@localhost) by albert.catwhisker.org (8.14.4/8.14.4/Submit) id o9TGwqHe017152 for freebsd-performance@freebsd.org; Fri, 29 Oct 2010 09:58:52 -0700 (PDT) (envelope-from david) Date: Fri, 29 Oct 2010 09:58:52 -0700 From: David Wolfskill To: freebsd-performance@freebsd.org Message-ID: <20101029165852.GD10729@albert.catwhisker.org> References: <20101020174854.GZ21226@albert.catwhisker.org> <4CBF8032.8000609@freebsd.org> <20101025165548.GE1519@albert.catwhisker.org> <4CC662A1.9030708@freebsd.org> <20101026112916.GB2262@albert.catwhisker.org> <201010261434.o9QEY32Q037618@lava.sentex.ca> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="YToU2i3Vx8H2dn7O" Content-Disposition: inline In-Reply-To: <201010261434.o9QEY32Q037618@lava.sentex.ca> User-Agent: Mutt/1.4.2.3i Subject: Re: Possible evidence of performance regression for 8.1-S (vs. 7.1) 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, 29 Oct 2010 16:58:53 -0000 --YToU2i3Vx8H2dn7O Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Tue, Oct 26, 2010 at 10:34:08AM -0400, Mike Tancsa wrote: > ... > ULE is the default in 7 as well. Perhaps remove some of the kernel=20 > options not in 7, that are in 8 by default? What is the disk=20 > subsystem ? just ata ? >=20 > They seem innocuous enough, but worth a try >=20 > options HWPMC_HOOKS # Necessary kernel hooks for hwpmc= (4) > options MAC # TrustedBSD MAC Framework > options FLOWTABLE # per-cpu routing cache OK; I now have the results from testing that. As noted earlier, we were using the MAC kernel config in 7.1, so I did not test removing the MAC option. I did build a kernel removing each of HWPMC_HOOKS and FLOWTABLE; results will be shown below as being associated with os "8.1-S-". To reduce the need for folks to find earlier messages to see the previous results, I'll paste them in, too: start stop real user sys host os = =20 1287525363 1287546846 21483.13 82628.20 21703.09 ref-7x 7.1-R+ = =20 1287548005 1287569100 21094.63 82853.19 22185.02 ref-7x 7.1-R+ = =20 1287570300 1287591371 21071.33 82756.81 21943.22 ref-7x 7.1-R+ = =20 1287592592 1287614103 21511.23 82637.30 21849.90 ref-7x 7.1-R+ = =20 1287615323 1287636770 21446.42 82715.81 21708.97 ref-7x 7.1-R+ = =20 = =20 1287436357 1287461948 25590.99 81502.22 18115.07 ref-8x 8.1-S@r214029= =20 1287462797 1287488766 25969.26 81452.14 17920.14 ref-8x 8.1-S@r214029= =20 1287489641 1287515287 25645.84 81548.40 18256.52 ref-8x 8.1-S@r214029= =20 1287516151 1287541481 25329.64 81546.23 18294.10 ref-8x 8.1-S@r214029= =20 1287542355 1287568599 26244.59 81431.47 17902.39 ref-8x 8.1-S@r214029= =20 = =20 1287710312 1287732046 21733.20 82688.01 22108.95 ref-8x 7.1-R+ = =20 1287733360 1287754549 21188.88 82869.09 21890.83 ref-8x 7.1-R+ = =20 1287755881 1287777566 21684.09 82772.50 21933.74 ref-8x 7.1-R+ = =20 1287879508 1287905173 25665.03 81696.22 18278.42 ref-8x 8.1-S[7.1-R+ = user] 1287906039 1287931709 25669.48 81735.00 18265.78 ref-8x 8.1-S[7.1-R+ = user] 1287932573 1287958275 25700.99 81700.40 18246.04 ref-8x 8.1-S[7.1-R+ = user] 1287959144 1287984859 25714.74 81800.40 18346.57 ref-8x 8.1-S[7.1-R+ = user] 1287985748 1288011752 26004.33 81569.28 17967.68 ref-8x 8.1-S[7.1-R+ = user] 1288198095 1288223776 25681.17 81550.12 18442.21 ref-8x 8.1-S- 1288224673 1288250350 25676.86 81561.61 18448.39 ref-8x 8.1-S- 1288269106 1288294837 25730.45 81535.38 18396.07 ref-8x 8.1-S- 1288295688 1288321338 25649.35 81552.34 18424.28 ref-8x 8.1-S- [Only 4 results posted, as one of the attempted builds failed.] And here are some ministat plots. As we have 3 "environment" variables (which machine is used, which kernel, and which userland), as well as 3 distinct variables being measured (real [wall clock] time; user CPU seconds; system CPU seconds), the file names reflect the measured variable ("real", "user", or "sys"), a "_h" to indicate that a "host designation" ("7" or "8") follows, a "_k" to indicate that a kernel designation follows ("7", "8", or "8-"), and a "_u" to indicate that a userland desgnation follows ("7" or "8"). So -- real time first: dwolf-bsd(8.1-S)[30] ministat -s real* x real_h7_k7_u7 + real_h8_k78_u7 * real_h8_k7_u7 : real_h8_k8-_u8 # real_h8_k8_u8 +--------------------------------------------------------------------------= ----+ |x * xx ** # #O@ O= #| ||___A_M| = | | |MA_| = | | |____A_M_| = | | |A = | | |___M_A___= _| | +--------------------------------------------------------------------------= ----+ N Min Max Median Avg Stddev x 5 21071.33 21511.23 21446.42 21321.348 218.96441 + 5 25665.03 26004.33 25700.99 25750.914 143.20282 Difference at 95.0% confidence 4429.57 +/- 269.817 20.7753% +/- 1.26548% (Student's t, pooled s =3D 185.003) * 3 21188.88 21733.2 21684.09 21535.39 301.08941 No difference proven at 95.0% confidence : 4 25649.35 25730.45 25679.015 25684.457 33.745994 Difference at 95.0% confidence 4363.11 +/- 264.927 20.4636% +/- 1.24254% (Student's t, pooled s =3D 166.989) # 5 25329.64 26244.59 25645.84 25756.064 355.43859 Difference at 95.0% confidence 4434.72 +/- 430.527 20.7994% +/- 2.01923% (Student's t, pooled s =3D 295.197) dwolf-bsd(8.1-S)[31]=20 User CPU: dwolf-bsd(8.1-S)[31] ministat -s user* x user_h7_k7_u7 + user_h8_k7_u7 * user_h8_k8-_u8 : user_h8_k8_u7 # user_h8_k8_u9 +--------------------------------------------------------------------------= ----+ | O = :| |x* O = :| |** #O# = :| |A| = | ||A = | | A = | | = A| | |A = | +--------------------------------------------------------------------------= ----+ N Min Max Median Avg Stddev x 5 21071.33 21511.23 21446.42 21321.348 218.96441 + 3 21188.88 21733.2 21684.09 21535.39 301.08941 No difference proven at 95.0% confidence * 4 25649.35 25730.45 25679.015 25684.457 33.745994 Difference at 95.0% confidence 4363.11 +/- 264.927 20.4636% +/- 1.24254% (Student's t, pooled s =3D 166.989) : 5 81569.28 81800.4 81700.4 81700.26 84.271865 Difference at 95.0% confidence 60378.9 +/- 241.959 283.185% +/- 1.13482% (Student's t, pooled s =3D 165.902) # 5 25329.64 26244.59 25645.84 25756.064 355.43859 Difference at 95.0% confidence 4434.72 +/- 430.527 20.7994% +/- 2.01923% (Student's t, pooled s =3D 295.197) dwolf-bsd(8.1-S)[32]=20 Finally, system CPU: dwolf-bsd(8.1-S)[32] ministat -s sys* x sys_h7_k7_u7 + sys_h8_k7_u7 * sys_h8_k8-_u8 : sys_h8_k8_u7 # sys_h8_k8_u8 +--------------------------------------------------------------------------= ----+ |# : ** x + = | |#: # ##:** x x+x= +x| | |__A__= _| | | |MA= _| | | AM = | | |__AM| = | ||___A__| = | +--------------------------------------------------------------------------= ----+ N Min Max Median Avg Stddev x 5 21703.09 22185.02 21849.9 21878.04 199.04094 + 3 21890.83 22108.95 21933.74 21977.84 115.55385 No difference proven at 95.0% confidence * 4 18396.07 18448.39 18433.245 18427.737 23.457507 Difference at 95.0% confidence -3450.3 +/- 239.945 -15.7706% +/- 1.09674% (Student's t, pooled s =3D 151.242) : 5 17967.68 18346.57 18265.78 18220.898 146.5163 Difference at 95.0% confidence -3657.14 +/- 254.882 -16.716% +/- 1.16501% (Student's t, pooled s =3D 174.763) # 5 17902.39 18294.1 18115.07 18097.644 182.87428 Difference at 95.0% confidence -3780.4 +/- 278.75 -17.2794% +/- 1.27411% (Student's t, pooled s =3D 191.129) dwolf-bsd(8.1-S)[33]=20 Peace, david --=20 David H. Wolfskill david@catwhisker.org Depriving a girl or boy of an opportunity for education is evil. See http://www.catwhisker.org/~david/publickey.gpg for my public key. --YToU2i3Vx8H2dn7O Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.16 (FreeBSD) iEYEARECAAYFAkzK/UsACgkQmprOCmdXAD3C/gCdGIuyT3XyRVUz68Ly33zFJGUf bVEAniOjHlWnBphcrvdd74Zgh1sGvrW7 =eCAu -----END PGP SIGNATURE----- --YToU2i3Vx8H2dn7O--