Re: reducing statistics write overhead

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Euler Taveira de Oliveira <euler(at)timbira(dot)com>
Cc: Martin Pihlak <martin(dot)pihlak(at)gmail(dot)com>, Pg Hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: reducing statistics write overhead
Date: 2008-09-06 00:24:37
Views: Raw Message | Whole Thread | Download mbox | Resend email
Lists: pgsql-hackers

Euler Taveira de Oliveira <euler(at)timbira(dot)com> writes:
> If you can't afford a 500 msec pgstat time, then you need to make it
> tunable. Another ideas are (i) turn on/off pgstat per table or database
> and (ii) make the pgstat time tunable per table or database. You can use
> the reloptions column to store these info. These workarounds are much
> simpler than that you proposed and they're almost for free.

For normal usage on-demand dumping would be a really good thing; it'd
cut the overhead of having stats on tremendously, especially for people
who don't really use 'em. The particular signaling proposed here is
bogus, but if Martin can make it work in a cleaner fashion I think it's
likely a good idea.

regards, tom lane

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Euler Taveira de Oliveira 2008-09-06 00:56:30 Re: PostgreSQL SSL problem
Previous Message Tom Lane 2008-09-06 00:10:47 Re: [PATCH] "\ef <function>" in psql