Skip site navigation (1) Skip section navigation (2)

Re: splitting data into multiple tables

From: Viji V Nair <viji(at)fedoraproject(dot)org>
To: Greg Smith <greg(at)2ndquadrant(dot)com>
Cc: nair rajiv <nair331(at)gmail(dot)com>, Andres Freund <andres(at)anarazel(dot)de>, pgsql-performance(at)postgresql(dot)org
Subject: Re: splitting data into multiple tables
Date: 2010-01-26 18:23:25
Message-ID: (view raw or whole thread)
Lists: pgsql-performance
On Tue, Jan 26, 2010 at 11:11 PM, Greg Smith <greg(at)2ndquadrant(dot)com> wrote:

> Viji V Nair wrote:
>> A 15k rpm SAS drive will give you a throughput of 12MB  and 120 IOPS. Now
>> you can calculate the number of disks, specifically spindles, for getting
>> your desired throughput and IOPs
> I think you mean 120MB/s for that first part.  Regardless, presuming you
> can provision a database just based on IOPS rarely works.  It's nearly
> impossible to estimate what you really need anyway for a database app, given
> that much of real-world behavior depends on the cached in memory vs.
> uncached footprint of the data you're working with.  By the time you put a
> number of disks into an array, throw a controller card cache on top of it,
> then add the OS and PostgreSQL caches on top of those, you are so far
> disconnected from the underlying drive IOPS that speaking in those terms
> doesn't get you very far.  I struggle with this every time I talk with a SAN
> vendor.  Their fixation on IOPS without considering things like how
> sequential scans mixed into random I/O will get handled is really
> disconnected from how databases work in practice.  For example, I constantly
> end up needing to detune IOPS in favor of readahead to make "SELECT x,y,z
> FROM t" run at an acceptable speed on big tables.
Yes, you are right.

There are catches in the SAN controllers also. SAN vendors wont give that
much information regarding their internal controller design. They will say
they have 4 external 4G ports, you should also check how many internal ports
they have and the how the controllers are operating,  in Active-Active or
Active- Standby mode.

> --
> Greg Smith    2ndQuadrant   Baltimore, MD
> PostgreSQL Training, Services and Support
> greg(at)2ndQuadrant(dot)com

In response to


pgsql-performance by date

Next:From: Robert HaasDate: 2010-01-26 19:35:07
Subject: Re: Poor query plan across OR operator
Previous:From: Richard NeillDate: 2010-01-26 17:41:32
Subject: Re: Should the optimiser convert a CASE into a WHERE if it can?

Privacy Policy | About PostgreSQL
Copyright © 1996-2015 The PostgreSQL Global Development Group