From: Wanpeng Li <liwanp@linux.vnet.ibm.com>
To: Bob Liu <lliubbo@gmail.com>
Cc: Linux-MM <linux-mm@kvack.org>,
Andrew Morton <akpm@linux-foundation.org>,
Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>,
Seth Jennings <sjenning@linux.vnet.ibm.com>,
Nitin Gupta <ngupta@vflare.org>,
bob.liu@oracle.com, Mel Gorman <mgorman@suse.de>
Subject: Re: Testing results of zswap
Date: Tue, 6 Aug 2013 19:37:49 +0800 [thread overview]
Message-ID: <20130806113749.GA14314@hacker.(null)> (raw)
In-Reply-To: <CAA_GA1fiEJYxqAZ1c0BneuftB5g8d+2_mYBj=4iE=1EcYaTx7w@mail.gmail.com>
[-- Attachment #1: Type: text/plain, Size: 626 bytes --]
On Thu, Jun 27, 2013 at 10:03:52AM +0800, Bob Liu wrote:
>Hi All,
>
>These days I have been testing zswap.
>I found that the total ram size of my testing machine effected the
>testing result.
>
>If I limit RAM size to 2G using "mem=", the performance of zswap is
>very disappointing,
>But if I use larger RAM size such as 8G, the performance is much better.
>Even with RAM size 8G, zswap will slow down the speed of parallelio.
>
>I run the testing(mmtest-0.10 with
>config-global-dhp__parallelio-memcachetest) after the default
>distribution booted every time.
>
Hi Bob,
I see improvement against v3.11-rc1 w/ 2G memory.
[-- Attachment #2: test --]
[-- Type: text/plain, Size: 2061 bytes --]
nozswap2 zswap2
nozswap2G zswap2G
Ops memcachetest-0M 12731.00 ( 0.00%) 11561.00 ( -9.19%)
Ops memcachetest-201M 11373.00 ( 0.00%) 11084.00 ( -2.54%)
Ops memcachetest-672M 11350.00 ( 0.00%) 10910.00 ( -3.88%)
Ops memcachetest-1142M 11057.00 ( 0.00%) 11060.00 ( 0.03%)
Ops io-duration-0M 0.00 ( 0.00%) 0.00 ( 0.00%)
Ops io-duration-201M 4.00 ( 0.00%) 5.00 (-25.00%)
Ops io-duration-672M 9.00 ( 0.00%) 7.00 ( 22.22%)
Ops io-duration-1142M 11.00 ( 0.00%) 13.00 (-18.18%)
Ops swaptotal-0M 451.00 ( 0.00%) 26208.00 (-5711.09%)
Ops swaptotal-201M 158775.00 ( 0.00%) 37548.00 ( 76.35%)
Ops swaptotal-672M 139599.00 ( 0.00%) 42514.00 ( 69.55%)
Ops swaptotal-1142M 137789.00 ( 0.00%) 34580.00 ( 74.90%)
Ops swapin-0M 451.00 ( 0.00%) 11830.00 (-2523.06%)
Ops swapin-201M 29082.00 ( 0.00%) 16169.00 ( 44.40%)
Ops swapin-672M 26611.00 ( 0.00%) 19574.00 ( 26.44%)
Ops swapin-1142M 27238.00 ( 0.00%) 15625.00 ( 42.64%)
Ops minorfaults-0M 557891.00 ( 0.00%) 575357.00 ( -3.13%)
Ops minorfaults-201M 743922.00 ( 0.00%) 595238.00 ( 19.99%)
Ops minorfaults-672M 727870.00 ( 0.00%) 653777.00 ( 10.18%)
Ops minorfaults-1142M 722946.00 ( 0.00%) 595093.00 ( 17.68%)
Ops majorfaults-0M 116.00 ( 0.00%) 4053.00 (-3393.97%)
Ops majorfaults-201M 4251.00 ( 0.00%) 4412.00 ( -3.79%)
Ops majorfaults-672M 3854.00 ( 0.00%) 4971.00 (-28.98%)
Ops majorfaults-1142M 3910.00 ( 0.00%) 4033.00 ( -3.15%)
next prev parent reply other threads:[~2013-08-06 11:38 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-06-27 2:03 Testing results of zswap Bob Liu
2013-07-15 2:56 ` Bob Liu
2013-07-15 14:56 ` Seth Jennings
2013-07-16 12:16 ` Bob Liu
2013-08-06 11:37 ` Wanpeng Li [this message]
2013-08-06 11:37 ` Wanpeng Li
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='20130806113749.GA14314@hacker.(null)' \
--to=liwanp@linux.vnet.ibm.com \
--cc=akpm@linux-foundation.org \
--cc=bob.liu@oracle.com \
--cc=konrad.wilk@oracle.com \
--cc=linux-mm@kvack.org \
--cc=lliubbo@gmail.com \
--cc=mgorman@suse.de \
--cc=ngupta@vflare.org \
--cc=sjenning@linux.vnet.ibm.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).