Re: Large Objects

From: Bruno Wolff III <bruno(at)wolff(dot)to>
To: Dan Boitnott <dan(at)mcneese(dot)edu>
Cc: pgsql-general(at)postgresql(dot)org
Subject: Re: Large Objects
Date: 2004-12-31 06:21:40
Message-ID: 20041231062140.GC17555@wolff.to
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

On Mon, Dec 27, 2004 at 10:39:48 -0600,
Dan Boitnott <dan(at)mcneese(dot)edu> wrote:
> I need to do some investigation into the way Postgres handles large
> objects for a major project involving large objects. My questions are:

I don't know the answer to all of your questions.

> * Is it practical/desirable to store files MIME-Encoded inside a
> text field?

This should be possible if the files aren't too large. bytea is another type
that might be better to use.

> * The obvious disadvantages:
> * slow, Slow, SLOW

If you always need to access the whole file this might not be too bad.
But if you only need to access a small part, you are going to pay a big
cost as the whole record will need to be retrieved before you can pick
out the part you want.

> * significant increase in per-file storage requirements

It might not be too bad as large records can be compressed. That should get
back some of the bloat from uuencoding.

In response to

Responses

Browse pgsql-general by date

  From Date Subject
Next Message Tatsuo Ishii 2004-12-31 06:55:37 Re: pg_dump and pgpool
Previous Message Jaime Casanova 2004-12-31 06:19:05 Re: [PATCHES] reqd patch