Re: [PERFORM] A Better External Sort?

From: "Zeugswetter Andreas DAZ SD" <ZeugswetterA(at)spardat(dot)at>
To: "Ron Peacetree" <rjpeace(at)earthlink(dot)net>, <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: [PERFORM] A Better External Sort?
Date: 2005-10-06 10:56:55
Message-ID: E1539E0ED7043848906A8FF995BDA5797F0067@m0143.s-mxs.net
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers


> Now I've asked for the quickest path to detailed
> understanding of the pg IO subsystem. The goal being to get
> more up to speed on its coding details. Certainly not to
> annoy you or anyone else.

Basically pg does random 8k (compile time blocksize) reads/writes only.
Bitmap and sequential scans read 8k blocks in order.
Only WAL does n x 8k writes with one system call.

pg relys on the OS readahead (== larger block IO) to do efficient IO.
Basically the pg scan performance should match a dd if=file of=/dev/null
bs=8k,
unless CPU bound.

Andreas

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Gaetano Mendola 2005-10-06 11:13:57 Outer where pushed down
Previous Message Andreas Pflug 2005-10-06 10:50:29 version dependent compilation