pg.ddx.io  pgsql-admin@postgresql.org mailing list archive  
help / color / mirror / Atom feed
From: Evan Reiser <evan.reiser@gmail.com>
To: pgsql-admin@postgresql.org
Subject: High IOWAIT times, low iops? Can't figure out what the bottleneck is...
Date: Tue, 3 Jul 2007 13:19:43 -0400
Message-ID: <5f33a0cb0707031019n7ffd4929p4b558304d47c10d@mail.gmail.com> (raw)

I was wondering if you guys have some suggested settings for our server, i
think we are not hardware limited but the configureation is set up
incorrectly.  For some reason our database seems to have trouble handling
5-10+ inserts per second which seems to be a pretty trivial load for this
hardware, we're seeing very high %iowait, this is a pretty typical output
for #iostat -m 5

avg-cpu:  %user   %nice %system %iowait  %steal   %idle
                0.97      0.00    0.82         89.69     0.00      8.52

Device:            tps    MB_read/s    MB_wrtn/s    MB_read    MB_wrtn
sda              85.40         0.18         0.72                 0
      3
sdb               0.00         0.00         0.00                  0
      0
sdc              30.43         0.00         0.76                 0
      3
sdd              93.91         0.71         0.12                 3
     0




sda = 2x320GB 7200rpm in RAID1 (operating system)
sdc = 2x150GB 10krpm in RAID1    (transaction log is on this array)
sdd = 6x150GB 10krpm in RAID 10 (database is on the array)

we're using ReiserFS on each of the arrays as the file system
OS is debian Etch x64

raid controller = 3ware 9650 12port - 256MB cache

8GB RAM, core 2 duo - quad core 2.4ghz 8MB L2 Cache

it would seem like the io subsystem is the limiting factor, but i feel like
we should be barely hitting a wall, you can see from the example its writing
< 2MB/s to the array.  If i try to copy a big file (1-2GB) over to the
array, the array hits 300MB/s sustained, it can burst write files < 256mb AT
800MB/S

Here's some of our settings


shared_buffers = 1024MB
temp_buffers = 32MB
max_prepared_transactions = 50
work_mem = 64MB
maintenance_work_mem = 256MB
max_stack_depth = 7MB
max_fsm_pages = 512000

vacuum_cost_delay = 20ms

fsync = on
wal_buffers = 1MB
checkpoint_segments = 32

effective_cache_size = 4128MB

If you guys have any suggestions it would be greatly appreciated

-Evan Reiser

view thread (2+ messages)  latest in thread

Message-ID: <5f33a0cb0707031019n7ffd4929p4b558304d47c10d@mail.gmail.com>
Permalink:  ../5f33a0cb0707031019n7ffd4929p4b558304d47c10d@mail.gmail.com/
Also on:    postgresql.org/message-id/5f33a0cb0707031019n7ffd4929p4b558304d47c10d@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-admin@postgresql.org
  Cc: evan.reiser@gmail.com
  Subject: Re: High IOWAIT times, low iops? Can't figure out what the bottleneck is...
  In-Reply-To: <5f33a0cb0707031019n7ffd4929p4b558304d47c10d@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