Re: slow queries, possibly disk io

From: Tom Lane
Subject: Re: slow queries, possibly disk io
Date: ,
Msg-id: 11908.1117550013@sss.pgh.pa.us
(view: Whole thread, Raw)
In response to: Re: slow queries, possibly disk io  (Josh Close)
List: pgsql-performance

Tree view

slow queries, possibly disk io  (Josh Close, )
 Re: slow queries, possibly disk io  (John Arbash Meinel, )
  Re: slow queries, possibly disk io  (Josh Close, )
   Re: slow queries, possibly disk io  (Simon Riggs, )
    TIP 9: the planner will ignore... & datatypes  ("Marc Mamin", )
     Re: TIP 9: the planner will ignore... & datatypes  (Alvaro Herrera, )
 Re: slow queries, possibly disk io  (Dawid Kuroczko, )
 Re: slow queries, possibly disk io  (Christopher Kings-Lynne, )
  Re: slow queries, possibly disk io  (Josh Close, )
 slow queries, possibly disk io  (Josh Close, )
  Re: slow queries, possibly disk io  (Tom Lane, )
   Re: slow queries, possibly disk io  (Josh Close, )
   Re: slow queries, possibly disk io  (Josh Close, )
   Re: slow queries, possibly disk io  (Rudi Starcevic, )
    Re: slow queries, possibly disk io  (Josh Close, )
     Re: slow queries, possibly disk io  (Tom Lane, )
     Re: slow queries, possibly disk io  (Josh Close, )
      Re: slow queries, possibly disk io  (Manfred Koizar, )

Josh Close <> writes:
> There is 2 gigs of mem in this server. Here are my current settings.

> max_connections = 100
> shared_buffers = 50000
> sort_mem = 4096
> vacuum_mem = 32768
> effective_cache_size = 450000

> Shared buffers is set to 10% of total mem. Effective cache size is 90% of mem.

Uh, shared_buffers and effective_cache_size are both measured in pages,
which are 8K apiece unless you built with a nondefault BLCKSZ.  So the
above calculations are off ...

> Is there anything that can be done to have postgres grab more from
> memory rather than disk?

It's not so much a matter of what Postgres will do as what the kernel
will do.  Check to see if there is some limit on how much memory the
kernel will set aside for disk buffers.  Plain old "top" will generally
tell you what is going on, though interpreting its output sometimes
requires some wizardry.

            regards, tom lane


pgsql-performance by date:

From: John A Meinel
Date:
Subject: Re: postgresql-8.0.1 performance tuning
From: Cosimo Streppone
Date:
Subject: Re: postgresql-8.0.1 performance tuning