pg.ddx.io  pgsql-performance@postgresql.org mailing list archive  
help / color / mirror / Atom feed
From: Rajesh Kumar Mallah <mallah.rajesh@gmail.com>
To: Kevin Grittner <Kevin.Grittner@wicourts.gov>
Cc: Yeb Havinga <yebhavinga@gmail.com>
Cc: pgsql-performance@postgresql.org
Subject: Re: sudden spurt in swap utilization (was:cpu bound postgresql setup.)
Date: Fri, 25 Jun 2010 20:44:41 +0530
Message-ID: <AANLkTimPhkr2nvuLo-fpBfUdN08KsDFTwzL9ubJP5VG2@mail.gmail.com> (raw)
In-Reply-To: <4C2477A20200002500032B03@gw.wicourts.gov>
References: <AANLkTinDBPQL-SjLaCr7AVgzouWbRrjwA6MAgBih_ciy@mail.gmail.com>
	<4C2492F9.2020706@gmail.com>
	<AANLkTikUDoXMF-7N7csuYPWIzn9gSKm_g4NnmjvAYu8N@mail.gmail.com>
	<4C2477A20200002500032B03@gw.wicourts.gov>

I changed shared_buffers from 10G to 4G ,
swap usage has almost become nil.

# free
             total       used       free     shared    buffers     cached
Mem:      32871276   24575824    8295452          0      11064   22167324
-/+ buffers/cache:    2397436   30473840
Swap:      4192912        352    4192560

I also observed that there was a huge IO wait and load spike initially
which gradually reduced to normal levels. Now things seems to be
fine. but real test shall be during business hours.

vmstat output:
http://pastebin.com/ygu8gUhS

the iowait now is very respectable < 10% and CPU is idling most of
the time.

# vmstat 10
procs -----------memory---------- ---swap-- -----io---- --system--
-----cpu------
 r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id wa st
 2  1    352 8482444  11336 22299100    1    0   450   208    0    0
50  6 39  5  0
 4  0    352 8393840  11352 22304484    0    0   480   163 9260 12717
32  4 62  3  0
 5  1    352 8474788  11360 22308980    0    0   304   445 8295 12358
28  4 67  2  0
 3  0    352 8370672  11376 22316676    0    0   648   158 8760 13214
38  4 55  3  0
11  0    352 8193824  11392 22323572    0    0   621   577 8800 13163
37  4 56  3  0
 2  0    352 8229012  11408 22326664    0    0   169   405 9588 13696
34  4 61  1  0
 6  1    352 8319176  11424 22333144    0    0   559   170 8830 12929
32  4 61  3  0

I shall also try pgtune in a while.



view thread (15+ messages)  latest in thread

Message-ID: <AANLkTimPhkr2nvuLo-fpBfUdN08KsDFTwzL9ubJP5VG2@mail.gmail.com>
Permalink:  ../AANLkTimPhkr2nvuLo-fpBfUdN08KsDFTwzL9ubJP5VG2@mail.gmail.com/
Also on:    postgresql.org/message-id/AANLkTimPhkr2nvuLo-fpBfUdN08KsDFTwzL9ubJP5VG2@mail.gmail.com

 · 

reply

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Reply to all the recipients using the --to and --cc options:
  reply via email

  To: pgsql-performance@postgresql.org
  Cc: mallah.rajesh@gmail.com, Kevin.Grittner@wicourts.gov, yebhavinga@gmail.com
  Subject: Re: sudden spurt in swap utilization (was:cpu bound postgresql setup.)
  In-Reply-To: <AANLkTimPhkr2nvuLo-fpBfUdN08KsDFTwzL9ubJP5VG2@mail.gmail.com>

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox