Re: Expected accuracy of planner statistics

From: Csaba Nagy <nagy(at)ecircle-ag(dot)com>
To: Jorge Godoy <jgodoy(at)gmail(dot)com>
Cc: Arturo Perez <aperez(at)hayesinc(dot)com>, Postgres general mailing list <pgsql-general(at)postgresql(dot)org>
Subject: Re: Expected accuracy of planner statistics
Date: 2006-09-29 12:34:01
Message-ID: 1159533241.3874.285.camel@coppola.muc.ecircle.de
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

> [snip] Having an expensive process run once in a while and setting this value also
> sounds interesting. [snip]

What about externalizing the statistics calculation ? I mean, would it
make sense to have for e.g. a WAL-fed standby which has an additional
process which keeps the statistics in sync based on the incoming WAL
records, and feeds back the stats to the master as soon as they change
significantly ? The standby would be able to crunch ALL the data, in
almost real time... with almost no overhead for the master. It would
require though another server... but I guess where analyze is a problem,
throwing another server at it is not a problem.

Cheers,
Csaba.

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Shane Ambler 2006-09-29 13:02:25 Re: Can i see server SQL commands ?
Previous Message Jorge Godoy 2006-09-29 12:04:59 Re: Expected accuracy of planner statistics