| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Date: Tue, 18 Jun 2013 19:50:37 +0800
From: Bob Liu <lliubbo@...il.com>
To: Andrew Morton <akpm@...ux-foundation.org>
Cc: Seth Jennings <sjenning@...ux.vnet.ibm.com>,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
Nitin Gupta <ngupta@...are.org>,
Minchan Kim <minchan@...nel.org>,
Konrad Rzeszutek Wilk <konrad.wilk@...cle.com>,
Dan Magenheimer <dan.magenheimer@...cle.com>,
Robert Jennings <rcj@...ux.vnet.ibm.com>,
Jenifer Hopper <jhopper@...ibm.com>,
Mel Gorman <mgorman@...e.de>,
Johannes Weiner <jweiner@...hat.com>,
Rik van Riel <riel@...hat.com>,
Larry Woodman <lwoodman@...hat.com>,
Benjamin Herrenschmidt <benh@...nel.crashing.org>,
Dave Hansen <dave@...1.net>, Joe Perches <joe@...ches.com>,
Joonsoo Kim <iamjoonsoo.kim@....com>,
Cody P Schafer <cody@...ux.vnet.ibm.com>,
Hugh Dickens <hughd@...gle.com>,
Paul Mackerras <paulus@...ba.org>,
Linux-MM <linux-mm@...ck.org>,
Linux-Kernel <linux-kernel@...r.kernel.org>
Subject: Re: [PATCHv13 3/4] zswap: add to mm/
>
> So the minor fault rate improved and everything else got worse?
I did the test again, in a new clean environment.
I'm sure the config files are the same except enabled zswap.
v3.10-rc4 v3.10-rc4
2G-parallio-zswapbas 2G-parallio-nozswap
Ops memcachetest-0M 819.00 ( 0.00%) 1041.00 ( 27.11%)
Ops memcachetest-198M 736.00 ( 0.00%) 973.00 ( 32.20%)
Ops memcachetest-430M 700.00 ( 0.00%) 892.00 ( 27.43%)
Ops memcachetest-661M 672.00 ( 0.00%) 819.00 ( 21.88%)
Ops memcachetest-893M 675.00 ( 0.00%) 775.00 ( 14.81%)
Ops memcachetest-1125M 665.00 ( 0.00%) 764.00 ( 14.89%)
Ops memcachetest-1356M 641.00 ( 0.00%) 749.00 ( 16.85%)
Ops io-duration-0M 0.00 ( 0.00%) 0.00 ( 0.00%)
Ops io-duration-198M 111.00 ( 0.00%) 21.00 ( 81.08%)
Ops io-duration-430M 125.00 ( 0.00%) 29.00 ( 76.80%)
Ops io-duration-661M 153.00 ( 0.00%) 34.00 ( 77.78%)
Ops io-duration-893M 118.00 ( 0.00%) 36.00 ( 69.49%)
Ops io-duration-1125M 142.00 ( 0.00%) 43.00 ( 69.72%)
Ops io-duration-1356M 156.00 ( 0.00%) 50.00 ( 67.95%)
Ops swaptotal-0M 462237.00 ( 0.00%) 469193.00 ( -1.50%)
Ops swaptotal-198M 490462.00 ( 0.00%) 496201.00 ( -1.17%)
Ops swaptotal-430M 500469.00 ( 0.00%) 520400.00 ( -3.98%)
Ops swaptotal-661M 506038.00 ( 0.00%) 538872.00 ( -6.49%)
Ops swaptotal-893M 514930.00 ( 0.00%) 522590.00 ( -1.49%)
Ops swaptotal-1125M 521010.00 ( 0.00%) 526934.00 ( -1.14%)
Ops swaptotal-1356M 513128.00 ( 0.00%) 525241.00 ( -2.36%)
Ops swapin-0M 246425.00 ( 0.00%) 251226.00 ( -1.95%)
Ops swapin-198M 266446.00 ( 0.00%) 236126.00 ( 11.38%)
Ops swapin-430M 271586.00 ( 0.00%) 265193.00 ( 2.35%)
Ops swapin-661M 263404.00 ( 0.00%) 280281.00 ( -6.41%)
Ops swapin-893M 263958.00 ( 0.00%) 263004.00 ( 0.36%)
Ops swapin-1125M 276149.00 ( 0.00%) 261962.00 ( 5.14%)
Ops swapin-1356M 264214.00 ( 0.00%) 262571.00 ( 0.62%)
Ops minorfaults-0M 629965.00 ( 0.00%) 625759.00 ( 0.67%)
Ops minorfaults-198M 641320.00 ( 0.00%) 769752.00 (-20.03%)
Ops minorfaults-430M 635270.00 ( 0.00%) 678590.00 ( -6.82%)
Ops minorfaults-661M 625123.00 ( 0.00%) 669308.00 ( -7.07%)
Ops minorfaults-893M 625165.00 ( 0.00%) 656343.00 ( -4.99%)
Ops minorfaults-1125M 624358.00 ( 0.00%) 657954.00 ( -5.38%)
Ops minorfaults-1356M 619724.00 ( 0.00%) 672035.00 ( -8.44%)
Ops majorfaults-0M 57664.00 ( 0.00%) 39395.00 ( 31.68%)
Ops majorfaults-198M 59747.00 ( 0.00%) 38758.00 ( 35.13%)
Ops majorfaults-430M 63453.00 ( 0.00%) 39819.00 ( 37.25%)
Ops majorfaults-661M 61310.00 ( 0.00%) 40171.00 ( 34.48%)
Ops majorfaults-893M 62544.00 ( 0.00%) 37576.00 ( 39.92%)
Ops majorfaults-1125M 60866.00 ( 0.00%) 36891.00 ( 39.39%)
Ops majorfaults-1356M 66598.00 ( 0.00%) 37123.00 ( 44.26%)
It shows clearly that swaptotal and minorfaults get improved a little
if enabled zswap.
But with the cost that the performance of everything else dropped a lot.
--
Regards,
--Bob
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists