Re: Is there a system of cache in pgsql 7.3rc1 ?

From: "scott(dot)marlowe" <scott(dot)marlowe(at)ihs(dot)com>
To: David Pradier <dpradier(at)apartia(dot)fr>
Cc: <pgsql-performance(at)postgresql(dot)org>
Subject: Re: Is there a system of cache in pgsql 7.3rc1 ?
Date: 2002-11-21 17:38:20
Message-ID: Pine.LNX.4.33.0211211035370.23081-100000@css120.ihs.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

On Thu, 21 Nov 2002, David Pradier wrote:

> Hi!
> This is my first post to the list.
>
> I'm currently searching to improve the performances of some requests,
> and recently switched to postgresql 7.3rc1.
>
> I thought there would be in this release a kind of cache for the results
> of requests (i.e. the second time a request is asked, if the concerned
> tables haven't changed, the result of the request isn't calculated another time, but
> taken from a cache.) ?
> Am i wrong ?
> Does such a mechanism exist ?
> Is it planned to create one ?

This issue has been discussed. The performance gains from a results cache
are not all that great, and postgresql's mvcc "locking" mechanism isn't a
good candidate to be served by results caching / updating. Generally
speaking, if you've got enough memory in your box, then the results are
"cached" in memory, requiring only sorting before being output.

This is a niche problem that is not likely to be implemented any time
soon.

In response to

Browse pgsql-performance by date

  From Date Subject
Next Message Rajesh Kumar Mallah. 2002-11-21 17:46:55 Re: H/W RAID 5 on slower disks versus no raid on faster HDDs
Previous Message scott.marlowe 2002-11-21 17:32:44 Re: vacuum full