Re: performance tuning: shared_buffers, sort_mem; swap

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: David Gilbert <dgilbert(at)velocet(dot)ca>
Cc: "Thomas O'Connell" <tfo(at)monsterlabs(dot)com>, pgsql-admin(at)postgresql(dot)org
Subject: Re: performance tuning: shared_buffers, sort_mem; swap
Date: 2002-08-14 13:41:01
Message-ID: 7558.1029332461@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-admin

David Gilbert <dgilbert(at)velocet(dot)ca> writes:
> "Tom" == Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> writes:
> Tom> Well, I must say this is the first time I've heard of "BSD-style
> Tom> shared memory".

> The canonical example is to mmap() a file multiply.

Oh. Yes, as Curt said, this issue has been beaten to death already;
see the archives.

You could in fact use a MAP_ANON region as a replacement for SysV shared
memory, but AFAICT you would lose a rather important interlock against
starting a new postmaster when old backends remain alive. Ugly as the
SysV API is, it does some things we need rather nicely.

> One bonus of bsd shared memory (when anonymous) is that it dies with
> the process ... even in the worst case.

But in fact we don't *want* it to be anonymous.

regards, tom lane

In response to

Responses

Browse pgsql-admin by date

  From Date Subject
Next Message Thomas O'Connell 2002-08-14 16:28:32 Re: performance tuning: shared_buffers, sort_mem; swap
Previous Message David Gilbert 2002-08-14 12:49:33 Re: performance tuning: shared_buffers, sort_mem; swap