Re: TOASTing smaller things

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: "Luke Lonergan" <llonergan(at)greenplum(dot)com>
Cc: "Bruce Momjian" <bruce(at)momjian(dot)us>, "Jan Wieck" <JanWieck(at)Yahoo(dot)com>, "Chris Browne" <cbbrowne(at)acm(dot)org>, pgsql-hackers(at)postgresql(dot)org
Subject: Re: TOASTing smaller things
Date: 2007-04-12 20:40:20
Message-ID: 5124.1176410420@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

"Luke Lonergan" <llonergan(at)greenplum(dot)com> writes:
> On 4/12/07 9:24 AM, "Bruce Momjian" <bruce(at)momjian(dot)us> wrote:
>> At this point I would be happy just to set the TOAST threshold to a
>> value defined as optimal, rather than as the most minimal use of TOAST
>> possible.

> I agree that's a good starting point, I guess I was thinking that was
> already included in the work that Tom has been doing.

No. I put in the code needed to decouple toast tuple size from toasting
threshold, but I don't have the time or interest to run performance
tests to see whether there are better default values than the historical
quarter-page values. Someone should do that before 8.3 beta ...

> If not, we can add a
> TODO like this as a precursor to the ones above:

> - Allow specification of TOAST size threshold (in bytes) on a per table
> basis

I would suggest that *all* of those TODOs are premature in the absence
of experimental evidence about the effect of varying the parameters.
If we end up finding out that the existing settings are about right
anyway across a range of test cases, who needs more knobs? We've got
too many mostly-useless knobs already.

regards, tom lane

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Steve 2007-04-12 21:03:17 Strangely Variable Query Performance
Previous Message Luke Lonergan 2007-04-12 20:35:25 Re: TOASTing smaller things