From mboxrd@z Thu Jan 1 00:00:00 1970 From: =?utf-8?B?SsO2cm4=?= Engel Subject: Re: Filesystem benchmarks on reasonably fast hardware Date: Thu, 21 Jul 2011 12:42:46 +0200 Message-ID: <20110721104246.GO1437@logfs.org> References: <20110717160501.GA1437@logfs.org> <20110719131958.GA9359@dastard> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: linux-fsdevel@vger.kernel.org To: Dave Chinner Return-path: Received: from [85.183.114.52] ([85.183.114.52]:35980 "EHLO Dublin.logfs.org" rhost-flags-FAIL-FAIL-OK-FAIL) by vger.kernel.org with ESMTP id S1751495Ab1GUKmR (ORCPT ); Thu, 21 Jul 2011 06:42:17 -0400 Content-Disposition: inline In-Reply-To: <20110719131958.GA9359@dastard> Sender: linux-fsdevel-owner@vger.kernel.org List-ID: On Tue, 19 July 2011 23:19:58 +1000, Dave Chinner wrote: > On Sun, Jul 17, 2011 at 06:05:01PM +0200, J=C3=B6rn Engel wrote: > > xfs: > > =3D=3D=3D=3D > ..... > > seqwr 1 2 4 8 16 32 64 128 > > 16384 39956 39695 39971 39913 37042 37538 36591 32179=09 > > 8192 67934 66073 30963 29038 29852 25210 23983 28272=09 > > 4096 89250 81417 28671 18685 12917 14870 22643 22237=09 > > 2048 140272 120588 140665 140012 137516 139183 131330 129684=09 > > 1024 217473 147899 210350 218526 219867 220120 219758 215166=09 > > 512 328260 181197 211131 263533 294009 298203 301698 298013=09 >=20 > OK, I can explain the pattern here where throughput drops off at 2-4 > threads. It's not as simple as the seqrd case, but it's related to > the fact that this workload is an append write workload. See the > patch description below for why that matters. >=20 > As it is, the numbers I get for 16k seqwr on my hardawre are as > follows: >=20 > seqwr 1 2 4 8 16 > vanilla 3072 2734 2506 not tested... > patched 2984 4156 4922 5175 5120 >=20 > Looks like my hardware is topping out at ~5-6kiops no matter the > block size here. Which, no matter how you look at it, is a > significant improvement. ;) My numbers include some regressions, although the improvements clearly dominate. Below is a diff (or div) between new kernel with both your patches applied and vanilla. >1 means improvement, <1 means regression. seqrd 1 2 4 8 16 32 64 128 16384 1.037 1.975 3.726 6.643 8.901 8.902 8.431 8.365=09 8192 1.015 1.871 3.459 6.424 11.457 12.829 13.542 13.490=09 4096 1.009 1.790 2.942 5.179 9.634 16.667 17.652 17.666=09 2048 1.005 1.709 2.525 4.196 7.479 14.022 22.032 22.100=09 1024 1.017 1.624 2.328 3.587 6.112 11.365 20.311 21.315=09 512 1.012 1.829 2.374 3.365 5.352 9.459 16.809 18.771=09 rndrd 1 2 4 8 16 32 64 128 16384 1.042 1.037 1.036 1.043 1.051 1.011 1.002 1.001=09 8192 1.020 1.020 1.028 1.040 1.057 1.064 1.002 1.001=09 4096 1.011 1.007 1.021 1.036 1.059 1.086 1.021 1.001=09 2048 1.002 1.010 1.018 1.025 1.057 1.100 1.098 1.003=09 1024 1.001 1.002 1.023 1.007 1.072 1.112 1.162 1.102=09 512 0.998 1.010 1.004 1.035 1.088 1.121 1.156 1.127=09 seqwr 1 2 4 8 16 32 64 128 16384 0.942 0.949 0.942 0.945 1.017 1.004 1.030 1.172=09 8192 1.144 1.177 2.517 2.687 2.611 3.091 3.246 2.741=09 4096 1.389 1.506 4.228 6.443 9.313 8.064 5.276 5.394=09 2048 1.139 1.278 1.080 1.076 1.094 1.087 1.142 1.148=09 1024 0.852 1.190 0.806 0.783 0.776 0.774 0.769 0.774=09 512 0.709 1.273 1.055 0.847 0.758 0.744 0.738 0.746=09 rndwr 1 2 4 8 16 32 64 128 16384 1.013 1.003 1.002 1.005 1.007 1.006 1.003 1.002=09 8192 1.023 1.005 1.007 1.006 1.006 1.004 1.004 1.006=09 4096 1.020 1.007 1.007 1.007 1.007 1.007 1.008 1.007=09 2048 0.901 1.017 1.007 1.008 1.008 1.009 1.008 1.007=09 1024 0.848 0.949 1.003 0.990 1.001 1.006 1.006 1.005=09 512 0.821 0.833 0.948 0.956 0.935 0.929 0.921 0.914=09 Raw results: seqrd 1 2 4 8 16 32 64 128 16384 4873 8738 16382 29241 39111 39152 39137 39140=09 8192 6326 10900 20054 37263 66391 78437 78449 78404=09 4096 9181 15816 26130 46073 85492 148172 157276 157329=09 2048 14995 24588 36009 59790 106685 200012 314373 315440=09 1024 24248 36841 51972 80207 136529 253175 451709 475353=09 512 37813 62164 79048 112175 178246 314959 558458 624534=09 rndrd 1 2 4 8 16 32 64 128 16384 4778 8554 14724 23507 33666 39065 39109 39104=09 8192 6152 11409 20862 35814 56123 75776 78370 78380=09 4096 8335 15643 29662 53953 91867 136643 157314 157325=09 2048 11973 22885 43474 81545 148087 239997 314198 315680=09 1024 16547 31345 61123 113283 222737 387234 562457 632767=09 512 20590 40134 73333 135621 294448 543117 793329 818861=09 seqwr 1 2 4 8 16 32 64 128 16384 37629 37651 37667 37711 37658 37674 37687 37727=09 8192 77691 77747 77948 78017 77940 77931 77847 77488=09 4096 123997 122607 121219 120394 120301 119908 119457 119939=09 2048 159816 154063 151987 150608 150449 151298 150016 148852=09 1024 185215 175977 169562 171078 170649 170420 169076 166614=09 512 232890 230669 222830 223140 222877 221812 222588 222369=09 rndwr 1 2 4 8 16 32 64 128 16384 38944 38256 38227 38312 38438 38432 38331 38313=09 8192 79773 77378 77453 77425 77473 77500 77458 77535=09 4096 163925 157167 158258 158192 158244 158281 158229 158252=09 2048 293295 322480 320206 321022 321375 321926 322298 322558=09 1024 368010 616516 652359 645514 654715 658132 659513 659125=09 512 411236 730015 1223437 1164632 1163705 1178235 1184450 1186594=09 J=C3=B6rn --=20 Ninety percent of everything is crap. -- Sturgeon's Law -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel= " in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html