From: Balbir Singh <balbir@linux.vnet.ibm.com>
To: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Ying Han <yinghan@google.com>,
"linux-mm@kvack.org" <linux-mm@kvack.org>,
"kosaki.motohiro@jp.fujitsu.com" <kosaki.motohiro@jp.fujitsu.com>,
"nishimura@mxp.nes.nec.co.jp" <nishimura@mxp.nes.nec.co.jp>,
"akpm@linux-foundation.org" <akpm@linux-foundation.org>,
Johannes Weiner <jweiner@redhat.com>,
"minchan.kim@gmail.com" <minchan.kim@gmail.com>,
Michal Hocko <mhocko@suse.cz>
Subject: Re: [PATCH 0/7] memcg background reclaim , yet another one.
Date: Mon, 2 May 2011 11:39:49 +0530 [thread overview]
Message-ID: <20110502060949.GN6547@balbir.in.ibm.com> (raw)
In-Reply-To: <20110425182529.c7c37bb4.kamezawa.hiroyu@jp.fujitsu.com>
* KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> [2011-04-25 18:25:29]:
>
> This patch is based on Ying Han's one....at its origin, but I changed too much ;)
> Then, start this as new thread.
>
> (*) This work is not related to the topic "rewriting global LRU using memcg"
> discussion, at all. This kind of hi/low watermark has been planned since
> memcg was born.
>
> At first, per-memcg background reclaim is used for
> - helping memory reclaim and avoid direct reclaim.
> - set a not-hard limit of memory usage.
>
> For example, assume a memcg has its hard-limit as 500M bytes.
> Then, set high-watermark as 400M. Here, memory usage can exceed 400M up to 500M
> but memory usage will be reduced automatically to 400M as time goes by.
>
> This is useful when a user want to limit memory usage to 400M but don't want to
> see big performance regression by hitting limit when memory usage spike happens.
>
> 1) == hard limit = 400M ==
> [root@rhel6-test hilow]# time cp ./tmpfile xxx
> real 0m7.353s
> user 0m0.009s
> sys 0m3.280s
>
What do the stats look like (graphed during this period?)
> 2) == hard limit 500M/ hi_watermark = 400M ==
> [root@rhel6-test hilow]# time cp ./tmpfile xxx
>
> real 0m6.421s
> user 0m0.059s
> sys 0m2.707s
>
What do the stats look like (graphed during this period?) for
comparison. Does the usage extend beyond 400 very often?
> Above is a brief result on VM and needs more study. But my impression is positive.
> I'd like to use bigger real machine in the next time.
>
> Here is a short list of updates from Ying Han's one.
>
> 1. use workqueue and visit memcg in round robin.
> 2. only allow setting hi watermark. low-watermark is automatically determined.
> This is good for avoiding bad cpu usage by background reclaim.
> 3. totally rewrite algorithm of shrink_mem_cgroup for round-robin.
> 4. fixed get_scan_count() , this was problematic.
> 5. added some statistics, which I think necessary.
> 6. added documenation
>
> Then, the algorithm is not a cut-n-paste from kswapd. I thought kswapd should be
> updated...and 'priority' in vmscan.c seems to be an enemy of memcg ;)
>
Thanks for looking into this.
--
Three Cheers,
Balbir
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
prev parent reply other threads:[~2011-05-02 6:10 UTC|newest]
Thread overview: 68+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-04-25 9:25 [PATCH 0/7] memcg background reclaim , yet another one KAMEZAWA Hiroyuki
2011-04-25 9:28 ` [PATCH 1/7] memcg: add high/low watermark to res_counter KAMEZAWA Hiroyuki
2011-04-26 17:54 ` Ying Han
2011-04-29 13:33 ` Michal Hocko
2011-05-01 6:06 ` KOSAKI Motohiro
2011-05-03 6:49 ` Michal Hocko
2011-05-03 7:45 ` KOSAKI Motohiro
2011-05-03 8:25 ` Michal Hocko
2011-05-03 17:01 ` Ying Han
2011-05-04 8:58 ` Michal Hocko
2011-05-04 17:16 ` Ying Han
2011-05-05 6:59 ` Michal Hocko
2011-05-06 5:28 ` KAMEZAWA Hiroyuki
2011-05-06 14:22 ` Johannes Weiner
2011-05-09 0:21 ` KAMEZAWA Hiroyuki
2011-05-09 5:47 ` Ying Han
2011-05-09 9:58 ` Johannes Weiner
2011-05-09 9:59 ` KAMEZAWA Hiroyuki
2011-05-10 4:43 ` Ying Han
2011-05-09 5:40 ` Ying Han
2011-05-09 7:10 ` KAMEZAWA Hiroyuki
2011-05-09 10:18 ` Johannes Weiner
2011-05-09 12:49 ` Michal Hocko
2011-05-09 23:49 ` KAMEZAWA Hiroyuki
2011-05-10 4:39 ` Ying Han
2011-05-10 4:51 ` Ying Han
2011-05-10 6:27 ` Johannes Weiner
2011-05-10 7:09 ` Ying Han
2011-05-04 3:55 ` KOSAKI Motohiro
2011-05-04 8:55 ` Michal Hocko
2011-05-09 3:24 ` KOSAKI Motohiro
2011-05-02 9:07 ` Balbir Singh
2011-05-06 5:30 ` KAMEZAWA Hiroyuki
2011-04-25 9:29 ` [PATCH 2/7] memcg high watermark interface KAMEZAWA Hiroyuki
2011-04-25 22:36 ` Ying Han
2011-04-25 9:31 ` [PATCH 3/7] memcg: select victim node in round robin KAMEZAWA Hiroyuki
2011-04-25 9:34 ` [PATCH 4/7] memcg fix scan ratio with small memcg KAMEZAWA Hiroyuki
2011-04-25 17:35 ` Ying Han
2011-04-26 1:43 ` KAMEZAWA Hiroyuki
2011-04-25 9:36 ` [PATCH 5/7] memcg bgreclaim core KAMEZAWA Hiroyuki
2011-04-26 4:59 ` Ying Han
2011-04-26 5:08 ` KAMEZAWA Hiroyuki
2011-04-26 23:15 ` Ying Han
2011-04-27 0:10 ` KAMEZAWA Hiroyuki
2011-04-27 1:01 ` KAMEZAWA Hiroyuki
2011-04-26 18:37 ` Ying Han
2011-04-25 9:40 ` [PATCH 6/7] memcg add zone_all_unreclaimable KAMEZAWA Hiroyuki
2011-04-25 9:42 ` [PATCH 7/7] memcg watermark reclaim workqueue KAMEZAWA Hiroyuki
2011-04-26 23:19 ` Ying Han
2011-04-27 0:31 ` KAMEZAWA Hiroyuki
2011-04-27 3:40 ` Ying Han
2011-04-25 9:43 ` [PATCH 8/7] memcg : reclaim statistics KAMEZAWA Hiroyuki
2011-04-26 5:35 ` Ying Han
2011-04-25 9:49 ` [PATCH 0/7] memcg background reclaim , yet another one KAMEZAWA Hiroyuki
2011-04-25 10:14 ` KAMEZAWA Hiroyuki
2011-04-25 22:21 ` Ying Han
2011-04-26 1:38 ` KAMEZAWA Hiroyuki
2011-04-26 7:19 ` Ying Han
2011-04-26 7:43 ` KAMEZAWA Hiroyuki
2011-04-26 8:43 ` Ying Han
2011-04-26 8:47 ` KAMEZAWA Hiroyuki
2011-04-26 23:08 ` Ying Han
2011-04-27 0:34 ` KAMEZAWA Hiroyuki
2011-04-27 1:19 ` Ying Han
2011-04-28 3:55 ` Ying Han
2011-04-28 4:05 ` KAMEZAWA Hiroyuki
2011-05-02 7:02 ` Balbir Singh
2011-05-02 6:09 ` Balbir Singh [this message]
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=20110502060949.GN6547@balbir.in.ibm.com \
--to=balbir@linux.vnet.ibm.com \
--cc=akpm@linux-foundation.org \
--cc=jweiner@redhat.com \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=kosaki.motohiro@jp.fujitsu.com \
--cc=linux-mm@kvack.org \
--cc=mhocko@suse.cz \
--cc=minchan.kim@gmail.com \
--cc=nishimura@mxp.nes.nec.co.jp \
--cc=yinghan@google.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).