Date: Tue, 28 May 2013 06:02:22 -0700 (PDT) From: Paul Pathiakis <pathiaki2@yahoo.com> To: Adrian Chadd <adrian@freebsd.org>, "O. Hartmann" <ohartman@zedat.fu-berlin.de> Cc: "freebsd-performance@freebsd.org" <freebsd-performance@freebsd.org> Subject: Re: New Phoronix performance benchmarks between some Linuxes and *BSDs Message-ID: <1369746142.64078.YahooMailNeo@web141401.mail.bf1.yahoo.com> In-Reply-To: <CAJ-VmokyRX5G%2B%2Bso=LJk5zEX56J5Q0R-Kiw7oqQJqKnLEMoZuw@mail.gmail.com> References: <20130528090822.6bfe8771@thor.walstatt.dyndns.org> <CAJ-VmokyRX5G%2B%2Bso=LJk5zEX56J5Q0R-Kiw7oqQJqKnLEMoZuw@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Outperform at "out of the box" testing. ;-)=0A=0ASo, if I have a "desktop" = distro like PCBSD, the only thing of relevance is putting up my own web ser= ver???? (Yes, the benchmark showed PCBSD seriously kicking butt with Apache= on static pages.... but why would I care on a desktop OS?)=0A=0APersonally= , I found the whole thing lacking coherency and relevancy on just about any= thing.=A0 =0A=0ADon't get me wrong, I do like the fact that this was done.= =A0 However, there are compiler differences (It was noted many times that C= LANG was used and it may have been a detriment but it doesn't go into the h= ow or why.) and other issues.=0A=0AThere was a benchmark on PostGreSQL, but= I didn't see any *BSD results?=0A=0ATransactions to a disk?=A0 Does this m= easure the "bundling" effect of the "groups of transactions" of ZFS?=A0 Tha= t's a whole lot less transactions that are sent to disk.=A0 (Does anyone kn= ow any place where this can be found?=A0 That is, how does the whole "bundl= ing of disk I/O" go from writing to memory, locking those writes, then send= ing all the info in one shot to the disk?=A0 This helps:=A0 http://blog.del= phix.com/ahl/2012/zfs-fundamentals-transaction-groups/ )=0A=0AI was working= at a company that had the intention of doing "electronic asset ingestion a= nd tagging".=A0 Basically, take any thing moved to the front end web server= s, copy it to disk, replicate it to other machines, etc... (maybe not in th= at order)=A0 The whole system was java based.=0A=0AThis was 3 years ago.=A0= I believe I was using Debian V4 (it had just come out....=A0 I don't recal= l the names etch, etc) and I took a single machine and rebuilt it 12 times:= =A0 OpenSuSe with ext2, ext3, xfs.=A0 Debian with ext2, ext3, xfs.=A0 CentO= S with ext2, ext3, xfs.=A0 FreeBSD 8.1 with ZFS, UFS2 w/ SU.=0A=0AWell, the= numbers came in and this was all done on the same HP 180 1u server rebuilt= that many times.=A0 I withheld the FBSD results as the development was don= e on Debian and people were "Linux inclined".=A0 The requisite was for 1500= 0 tpm per machine for I/O.=A0 Linux could only get to 3500.=A0 People were = pissed and they were looking at 5 years and $20m in time and development.= =A0 That's when I put the FBSD results in front of them..... 75,200 tpm.=A0= Now, this was THEIR measurements and THEIR benchmarks (The Engineering tea= m).=A0 The machine was doing nothing but running flat out on a horrible met= hod of using directory structure to organize the asset tags... (yeah, ugly)= =A0 However, ZFS almost didn't care compared to a traditional filesystem.= =A0 =0A=0ASo, what it comes down do is simple.... you can benchmark anythin= g you want with various "authoritative" benchmarks, but in the end, your be= nchmark on your data set (aka real world in your world) is the only thing t= hat matters.=0A=0ABTW, what happened in the situation I described?=A0 Despi= te, a huge cost savings and incredible performance....=A0 "We have to use D= ebian as we never put any type of automation in place that would allow us t= o be able to move from one OS to another"...=A0 Yeah, I guess a Systems Arc= hitect (like me) is something that people tend to overlook.=A0 System autom= ation to allow nimble transitions like that are totally overlooked.=0A=0ABe= nchmarks are "nice".=A0 However, tuning and understanding the underlying te= ch and what's it's good for is priceless.=A0 Knowing there are memory manag= ement issues, scheduling issues, certain types of I/O on certain FS that ca= use it to sing or sob, these are the things that will make someone invaluab= le.=A0 No one should be a tech bigot.=A0 The mantra should be:=A0 "The best= tech for the situation".=A0 No one should care if it's BSD, Linux, or Wind= oze if it's what works best in the situation.=0A=0AP=0A=0APS -=A0 When I se= e how many people are clueless about how much tech is ripped off from BSD t= o make other vendors' products just work and then they slap at BSD.... it's= pretty bad.=A0 GPLv3?=A0 Thank you... there are so many people going to a = "no GPL products in house" policy that there is a steady increase in BSD an= d ZFS.=A0 I can only hope GPLv4 becomes "If you use our stuff, we own all t= he machines and code that our stuff coexists on" :-)=0A=0A=0A=0A=0A=0A=0A__= ______________________________=0A From: Adrian Chadd <adrian@freebsd.org>= =0ATo: O. Hartmann <ohartman@zedat.fu-berlin.de> =0ACc: freebsd-performance= @freebsd.org =0ASent: Tuesday, May 28, 2013 5:03 AM=0ASubject: Re: New Phor= onix performance benchmarks between some Linuxes and *BSDs=0A =0A=0Aoutperf= orm at what?=0A=0A=0A=0Aadrian=0A=0AOn 28 May 2013 00:08, O. Hartmann <ohar= tman@zedat.fu-berlin.de> wrote:=0A> Phoronix has emitted another of its "fa= mous" performance tests=0A> comparing different flavours of Linux (their ob= vious favorite OS):=0A>=0A> http://www.phoronix.com/scan.php?page=3Darticle= &item=3Dbsd_linux_8way&num=3D1=0A>=0A> It is "impressive, too, to see that = PHORONIX did not benchmark the=0A> gaming performance - this is done exclus= ively on the Linux=0A> distributions, I guess in the lack of suitable graph= ics cards at=0A> Phronix (although it should be possible to compare the nVi= dia BLOB=0A> performance between each system).=0A>=0A> Although I'm not muc= h impressed by the way the benchmarks are=0A> orchestrated, Phoronix is the= only platform known to me providing those=0A> from time to time benchmarks= on most recent available operating systems.=0A>=0A> Also, the bad performa= nce of ZFS compared to to UFS2 seems to have a=0A> very harsh impact on sys= tems were that memory- and performance-hog ZFS=0A> isn't really needed.=0A>= =0A> Surprised and really disappointing (especially for me personally) is= =0A> the worse performance of the Rodinia benchmark on the BSDs, for what I= =0A> try to have deeper look inside to understand the circumstances of the= =0A> setups and what this scientific benchmark is supposed to do and=0A> me= asure.=0A>=0A> But the overall conclusion shown on Phoronix is that what I = see at our=0A> department which utilizes some Linux flavours, Ubuntu 12.01 = or Suse and=0A> in a majority CentOS (older versions), which all outperform= the several=0A> FreeBSd servers I maintain (FreeBSD 9.1-STABLE and FreeBSD= =0A> 10.0-CURRENT, so to end software compared to some older Linux kernels)= .=0A> _______________________________________________=0A> freebsd-performan= ce@freebsd.org mailing list=0A> http://lists.freebsd.org/mailman/listinfo/f= reebsd-performance=0A> To unsubscribe, send any mail to "freebsd-performanc= e-unsubscribe@freebsd.org"=0A______________________________________________= _=0Afreebsd-performance@freebsd.org mailing list=0Ahttp://lists.freebsd.org= /mailman/listinfo/freebsd-performance=0ATo unsubscribe, send any mail to "f= reebsd-performance-unsubscribe@freebsd.org" From owner-freebsd-performance@FreeBSD.ORG Tue May 28 14:28:08 2013 Return-Path: <owner-freebsd-performance@FreeBSD.ORG> Delivered-To: freebsd-performance@freebsd.org Received: from mx1.freebsd.org (mx1.FreeBSD.org [8.8.178.115]) by hub.freebsd.org (Postfix) with ESMTP id 2D4D9DD9; Tue, 28 May 2013 14:28:08 +0000 (UTC) (envelope-from db@nipsi.de) Received: from fop.bsdsystems.de (mx.bsdsystems.de [88.198.57.43]) by mx1.freebsd.org (Postfix) with ESMTP id ED873EE9; Tue, 28 May 2013 14:28:06 +0000 (UTC) Received: from hamstedm247370.global.intra.guj.com (unknown [194.12.218.135]) (using TLSv1 with cipher AES128-SHA (128/128 bits)) (No client certificate requested) by fop.bsdsystems.de (Postfix) with ESMTP id 8131B55D48; Tue, 28 May 2013 16:27:58 +0200 (CEST) Subject: Re: New Phoronix performance benchmarks between some Linuxes and *BSDs Mime-Version: 1.0 (Apple Message framework v1085) From: dennis berger <db@nipsi.de> In-Reply-To: <1369746142.64078.YahooMailNeo@web141401.mail.bf1.yahoo.com> Date: Tue, 28 May 2013 16:27:58 +0200 Message-Id: <F2325751-7571-44AB-8B84-C7BD76D4812F@nipsi.de> References: <20130528090822.6bfe8771@thor.walstatt.dyndns.org> <CAJ-VmokyRX5G++so=LJk5zEX56J5Q0R-Kiw7oqQJqKnLEMoZuw@mail.gmail.com> <1369746142.64078.YahooMailNeo@web141401.mail.bf1.yahoo.com> To: Paul Pathiakis <pathiaki2@yahoo.com> X-Mailer: Apple Mail (2.1085) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: quoted-printable X-Content-Filtered-By: Mailman/MimeDel 2.1.14 Cc: Adrian Chadd <adrian@freebsd.org>, "O. Hartmann" <ohartman@zedat.fu-berlin.de>, "freebsd-performance@freebsd.org" <freebsd-performance@freebsd.org> X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.14 Precedence: list List-Id: Performance/tuning <freebsd-performance.freebsd.org> List-Unsubscribe: <http://lists.freebsd.org/mailman/options/freebsd-performance>, <mailto:freebsd-performance-request@freebsd.org?subject=unsubscribe> List-Archive: <http://lists.freebsd.org/pipermail/freebsd-performance>; List-Post: <mailto:freebsd-performance@freebsd.org> List-Help: <mailto:freebsd-performance-request@freebsd.org?subject=help> List-Subscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-performance>, <mailto:freebsd-performance-request@freebsd.org?subject=subscribe> X-List-Received-Date: Tue, 28 May 2013 14:28:08 -0000 Hi, for me it's unknown what 100 TPS means in that particular case. But this = doesn't make sense at all and I don't see such a low number in the = postmark output here. I think I get around 4690+-435 IOPS with 95% confidence. Guest and the actual test system is FreeBSD9.1/64bit inside of = Virtualbox. Host system is MacOSX on 4year old macbook Storage is VDI file backed on a SSD (OCZ vortex 2) with a 2gb ZFS pool=20= When you I postmark with 25K transactions I get an output like this. = (http://fsbench.filesystems.org/bench/postmark-1_5.c) pm>run Creating files...Done Performing transactions..........Done Deleting files...Done Time: 6 seconds total 5 seconds of transactions (5000 per second) Files: 13067 created (2177 per second) Creation alone: 500 files (500 per second) Mixed with transactions: 12567 files (2513 per second) 12420 read (2484 per second) 12469 appended (2493 per second) 13067 deleted (2177 per second) Deletion alone: 634 files (634 per second) Mixed with transactions: 12433 files (2486 per second) Data: 80.71 megabytes read (13.45 megabytes per second) 84.59 megabytes written (14.10 megabytes per second) I did this 100 times on my notebook and summed up this. root@freedb:/pool/nase # ministat -n *.txt x alltransactions.txt + appended-no.txt * created-no.txt % deleted-no.txt # reed-no.txt N Min Max Median Avg = Stddev x 100 3571 5000 5000 4690.25 = 435.65125 + 100 1781 2493 2493 2338.84 = 216.8531 * 100 1633 2613 2613 2396.59 = 256.53752 % 100 1633 2613 2613 2396.59 = 256.53752 # 100 1774 2484 2484 2330.22 = 216.3084 When I check "zpool iostat 1" I see root@freedb:/pool/nase # zpool iostat 1 capacity operations bandwidth pool alloc free read write read write ---------- ----- ----- ----- ----- ----- ----- pool 10.6M 1.97G 0 8 28 312K ---------- ----- ----- ----- ----- ----- ----- pool 10.6M 1.97G 0 33 0 4.09M ---------- ----- ----- ----- ----- ----- ----- pool 10.6M 1.97G 0 0 0 0 ---------- ----- ----- ----- ----- ----- ----- pool 10.6M 1.97G 0 0 0 0 ---------- ----- ----- ----- ----- ----- ----- pool 10.6M 1.97G 0 0 0 0 ---------- ----- ----- ----- ----- ----- ----- pool 19.6M 1.97G 0 89 0 4.52M ---------- ----- ----- ----- ----- ----- ----- around 30-90 TPS bursts.=20 Did they counted this instead? -dennis Am 28.05.2013 um 15:02 schrieb Paul Pathiakis: > Outperform at "out of the box" testing. ;-) >=20 > So, if I have a "desktop" distro like PCBSD, the only thing of = relevance is putting up my own web server???? (Yes, the benchmark showed = PCBSD seriously kicking butt with Apache on static pages.... but why = would I care on a desktop OS?) >=20 > Personally, I found the whole thing lacking coherency and relevancy on = just about anything. =20 >=20 > Don't get me wrong, I do like the fact that this was done. However, = there are compiler differences (It was noted many times that CLANG was = used and it may have been a detriment but it doesn't go into the how or = why.) and other issues. >=20 > There was a benchmark on PostGreSQL, but I didn't see any *BSD = results? >=20 > Transactions to a disk? Does this measure the "bundling" effect of = the "groups of transactions" of ZFS? That's a whole lot less = transactions that are sent to disk. (Does anyone know any place where = this can be found? That is, how does the whole "bundling of disk I/O" = go from writing to memory, locking those writes, then sending all the = info in one shot to the disk? This helps: = http://blog.delphix.com/ahl/2012/zfs-fundamentals-transaction-groups/ ) >=20 > I was working at a company that had the intention of doing "electronic = asset ingestion and tagging". Basically, take any thing moved to the = front end web servers, copy it to disk, replicate it to other machines, = etc... (maybe not in that order) The whole system was java based. >=20 > This was 3 years ago. I believe I was using Debian V4 (it had just = come out.... I don't recall the names etch, etc) and I took a single = machine and rebuilt it 12 times: OpenSuSe with ext2, ext3, xfs. Debian = with ext2, ext3, xfs. CentOS with ext2, ext3, xfs. FreeBSD 8.1 with = ZFS, UFS2 w/ SU. >=20 > Well, the numbers came in and this was all done on the same HP 180 1u = server rebuilt that many times. I withheld the FBSD results as the = development was done on Debian and people were "Linux inclined". The = requisite was for 15000 tpm per machine for I/O. Linux could only get = to 3500. People were pissed and they were looking at 5 years and $20m = in time and development. That's when I put the FBSD results in front of = them..... 75,200 tpm. Now, this was THEIR measurements and THEIR = benchmarks (The Engineering team). The machine was doing nothing but = running flat out on a horrible method of using directory structure to = organize the asset tags... (yeah, ugly) However, ZFS almost didn't care = compared to a traditional filesystem. =20 >=20 > So, what it comes down do is simple.... you can benchmark anything you = want with various "authoritative" benchmarks, but in the end, your = benchmark on your data set (aka real world in your world) is the only = thing that matters. >=20 > BTW, what happened in the situation I described? Despite, a huge cost = savings and incredible performance.... "We have to use Debian as we = never put any type of automation in place that would allow us to be able = to move from one OS to another"... Yeah, I guess a Systems Architect = (like me) is something that people tend to overlook. System automation = to allow nimble transitions like that are totally overlooked. >=20 > Benchmarks are "nice". However, tuning and understanding the = underlying tech and what's it's good for is priceless. Knowing there = are memory management issues, scheduling issues, certain types of I/O on = certain FS that cause it to sing or sob, these are the things that will = make someone invaluable. No one should be a tech bigot. The mantra = should be: "The best tech for the situation". No one should care if = it's BSD, Linux, or Windoze if it's what works best in the situation. >=20 > P >=20 > PS - When I see how many people are clueless about how much tech is = ripped off from BSD to make other vendors' products just work and then = they slap at BSD.... it's pretty bad. GPLv3? Thank you... there are so = many people going to a "no GPL products in house" policy that there is a = steady increase in BSD and ZFS. I can only hope GPLv4 becomes "If you = use our stuff, we own all the machines and code that our stuff coexists = on" :-) >=20 >=20 >=20 >=20 >=20 >=20 > ________________________________ > From: Adrian Chadd <adrian@freebsd.org> > To: O. Hartmann <ohartman@zedat.fu-berlin.de>=20 > Cc: freebsd-performance@freebsd.org=20 > Sent: Tuesday, May 28, 2013 5:03 AM > Subject: Re: New Phoronix performance benchmarks between some Linuxes = and *BSDs >=20 >=20 > outperform at what? >=20 >=20 >=20 > adrian >=20 > On 28 May 2013 00:08, O. Hartmann <ohartman@zedat.fu-berlin.de> wrote: >> Phoronix has emitted another of its "famous" performance tests >> comparing different flavours of Linux (their obvious favorite OS): >>=20 >> = http://www.phoronix.com/scan.php?page=3Darticle&item=3Dbsd_linux_8way&num=3D= 1 >>=20 >> It is "impressive, too, to see that PHORONIX did not benchmark the >> gaming performance - this is done exclusively on the Linux >> distributions, I guess in the lack of suitable graphics cards at >> Phronix (although it should be possible to compare the nVidia BLOB >> performance between each system). >>=20 >> Although I'm not much impressed by the way the benchmarks are >> orchestrated, Phoronix is the only platform known to me providing = those >> from time to time benchmarks on most recent available operating = systems. >>=20 >> Also, the bad performance of ZFS compared to to UFS2 seems to have a >> very harsh impact on systems were that memory- and performance-hog = ZFS >> isn't really needed. >>=20 >> Surprised and really disappointing (especially for me personally) is >> the worse performance of the Rodinia benchmark on the BSDs, for what = I >> try to have deeper look inside to understand the circumstances of the >> setups and what this scientific benchmark is supposed to do and >> measure. >>=20 >> But the overall conclusion shown on Phoronix is that what I see at = our >> department which utilizes some Linux flavours, Ubuntu 12.01 or Suse = and >> in a majority CentOS (older versions), which all outperform the = several >> FreeBSd servers I maintain (FreeBSD 9.1-STABLE and FreeBSD >> 10.0-CURRENT, so to end software compared to some older Linux = kernels). >> _______________________________________________ >> 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" > _______________________________________________ > 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"
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1369746142.64078.YahooMailNeo>