Re: Question about memory allocations

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Ron <rjpeace(at)earthlink(dot)net>
Cc: Steve <cheetah(at)tanabi(dot)org>, pgsql-performance(at)postgresql(dot)org
Subject: Re: Question about memory allocations
Date: 2007-04-14 16:37:54
Message-ID: 4918.1176568674@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

Ron <rjpeace(at)earthlink(dot)net> writes:
> One of the reasons for the wide variance in suggested values for pg
> memory use is that pg 7.x and pg 8.x are =very= different beasts.
> If you break the advice into pg 7.x and pg 8.x categories, you find
> that there is far less variation in the suggestions.
> Bottom line: pg 7.x could not take advantage of larger sums of memory
> anywhere near as well as pg 8.x can.

Actually I think it was 8.1 that really broke the barrier in terms of
scalability of shared_buffers. Pre-8.1, the buffer manager just didn't
scale well enough to make it useful to use more than a few hundred meg.
(In fact, we never even bothered to fix the shared-memory-sizing
calculations to be able to deal with >2GB shared memory until 8.1;
if you try it in 8.0 it'll probably just crash.)

regards, tom lane

In response to

Browse pgsql-performance by date

  From Date Subject
Next Message Tom Lane 2007-04-14 17:19:33 Re: [HACKERS] choose_bitmap_and again (was Re: Strangely Variable Query Performance)
Previous Message Tom Lane 2007-04-14 15:52:45 Re: [HACKERS] choose_bitmap_and again (was Re: Strangely Variable Query Performance)