Re: Performance Issues

From: Dave Cramer <pg(at)fastcrypt(dot)com>
To: Richard Huxton <dev(at)archonet(dot)com>
Cc: Dhanaraj M <Dhanaraj(dot)M(at)Sun(dot)COM>, pgsql-hackers(at)postgresql(dot)org
Subject: Re: Performance Issues
Date: 2006-05-23 14:42:20
Message-ID: 77B03FE1-0B4E-47D7-8B4D-0CFB2DDC7783@fastcrypt.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers


On 23-May-06, at 10:24 AM, Richard Huxton wrote:

> Dhanaraj M wrote:
>> I have the following doubts.
>> 1. Does postgres create an index on every primary key? Usually,
>> queries are performed against a table on the primary key, so, an
>> index on it will be very useful.
>
> Yes, a unique index is used to enforce the primary-key.
>
>> 2. If 'm executing a complex query and it takes 10 seconds to
>> return the results -- it takes 10 seconds to execute the next time
>> also. I'm wondering if there's any kind of caching that can be
>> enabled -- so, the next time it takes <10 seconds to return the
>> results.
>
> Not of query results. Obviously data itself might be cached. You
> might want to look at memcached for this sort of thing.

Postgresql relies on the kernel buffers, and shared buffers for caching.

As someone else said postgresql is quite conservative when shipped.
Tuning helps considerably

Dave
>
> --
> Richard Huxton
> Archonet Ltd
>
> ---------------------------(end of
> broadcast)---------------------------
> TIP 2: Don't 'kill -9' the postmaster
>

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Simon Riggs 2006-05-23 14:49:15 Re: [ADMIN] does wal archiving block the current client
Previous Message Csaba Nagy 2006-05-23 14:29:30 Re: error-free disabling of individual child partition