Re: newbie design question re impact of VACUUM - Mailing list pgsql-general

From Richard Huxton
Subject Re: newbie design question re impact of VACUUM
Date
Msg-id 43720E83.1010809@archonet.com
Whole thread Raw
In response to newbie design question re impact of VACUUM  ("shakahshakah@gmail.com" <shakahshakah@gmail.com>)
List pgsql-general
shakahshakah@gmail.com wrote:
> After looking at "Chapter 22. Routine Database Maintenance Tasks"
> (http://www.postgresql.org/docs/8.1/interactive/maintenance.html), I
> started wondering about what (if any) consideration to give to to
> VACUUM issues in the following context.
>
> As a background, I'll be using Postgres in part as a processing queue
> for a 40-column stream of information (~ 250 bytes/row) with a
> sustained input rate of 20 rows/sec. This queue will be processed
> periodically (every few minutes), design constraints are to (1) only
> process each row once, and (2) keep the processed rows around for a
> period of time (say a month or so).

I would have the application vacuum the queue table after processing a
batch of rows. That's about as small as you could keep the table's disk
usage.

You are right in that updating the row will basically create a new
version of the row.

--
   Richard Huxton
   Archonet Ltd

pgsql-general by date:

Previous
From: "A. Kretschmer"
Date:
Subject: Re: How to obtain the code af a function ?
Next
From: Richard Huxton
Date:
Subject: Re: temporary tables