Re: Need for speed

From: Tom Lane
Subject: Re: Need for speed
Date: ,
Msg-id: 18975.1124290368@sss.pgh.pa.us
(view: Whole thread, Raw)
In response to: Re: Need for speed  (Ulrich Wisser)
List: pgsql-performance

Tree view

Need for speed  (Ulrich Wisser, )
 Re: Need for speed  (Richard Huxton, )
 Re: Need for speed  (John A Meinel, )
 Re: Need for speed  ("Jeffrey W. Baker", )
 Re: Need for speed  (Alex Turner, )
  Re: Need for speed  ("Jim C. Nasby", )
 Re: Need for speed  (Dennis Bjorklund, )
  Re: Need for speed  (Ulrich Wisser, )
   Re: Need for speed  (Tom Lane, )
   Re: Need for speed  ("Jeffrey W. Baker", )
   Re: Need for speed  (Josh Berkus, )
   Re: Need for speed  (Ron, )
 Re: Need for speed  (Ron, )
  Re: Need for speed  (Matthew Nuzum, )
   Need for speed 2  (Ulrich Wisser, )
    Re: Need for speed 2  (Frank Wiles, )
    Re: Need for speed 2  (Ron, )
     Re: Need for speed 2  (Kelly Burkhart, )
      Re: Need for speed 2  (Alex Turner, )
    Re: Need for speed 2  ("Merlin Moncure", )
     What *_mem to increase when running CLUSTER  (Andrew Lazarus, )
      Re: What *_mem to increase when running CLUSTER  (Steve Poe, )
      Re: What *_mem to increase when running CLUSTER  (Tom Lane, )
 Re: Need for speed  ("Roger Hand", )
  Re: Need for speed  (Christopher Browne, )

Ulrich Wisser <> writes:
> My machine has 2GB memory, please find postgresql.conf below.

> max_fsm_pages = 50000           # min max_fsm_relations*16, 6 bytes each

FWIW, that index I've been groveling through in connection with your
other problem contains an astonishingly large amount of dead space ---
almost 50%.  I suspect that you need a much larger max_fsm_pages
setting, and possibly more-frequent vacuuming, in order to keep a lid
on the amount of wasted space.

            regards, tom lane


pgsql-performance by date:

From: Ron
Date:
Subject: Re: Need for speed
From: "Merlin Moncure"
Date:
Subject: bitmap scan issues 8.1 devel