Skip site navigation (1) Skip section navigation (2)

Re: [PERFORM] A Better External Sort?

From: Josh Berkus <josh(at)agliodbs(dot)com>
To: Ron Peacetree <rjpeace(at)earthlink(dot)net>
Cc: Dann Corbit <DCorbit(at)connx(dot)com>, pgsql-hackers(at)postgresql(dot)org,pgsql-performance(at)postgresql(dot)org
Subject: Re: [PERFORM] A Better External Sort?
Date: 2005-09-27 16:15:06
Message-ID: 4339700A.2030803@agliodbs.com (view raw or flat)
Thread:
Lists: pgsql-hackerspgsql-performance
Ron,

I've somehow missed part of this thread, which is a shame since this is 
an area of primary concern for me.

Your suggested algorithm seems to be designed to relieve I/O load by 
making more use of the CPU.   (if I followed it correctly).  However, 
that's not PostgreSQL's problem; currently for us external sort is a 
*CPU-bound* operation, half of which is value comparisons.  (oprofiles 
available if anyone cares)

So we need to look, instead, at algorithms which make better use of 
work_mem to lower CPU activity, possibly even at the expense of I/O.

--Josh Berkus

In response to

pgsql-performance by date

Next:From: DarioDate: 2005-09-27 16:21:07
Subject: Re: VACUUM FULL vs CLUSTER
Previous:From: Kevin GrittnerDate: 2005-09-27 15:21:37
Subject: Re: [PERFORM] A Better External Sort?

pgsql-hackers by date

Next:From: Peter EisentrautDate: 2005-09-27 16:16:07
Subject: Re: Making pgxs builds work with a relocated installation
Previous:From: Jim C. NasbyDate: 2005-09-27 15:59:52
Subject: Re: Database file compatability

Privacy Policy | About PostgreSQL
Copyright © 1996-2014 The PostgreSQL Global Development Group