Re: 8.0.X and the ARC patent
From:
Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To:
Michael Adler <adler(at)pobox(dot)com>
Cc:
PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org>,Mark Wong <markw(at)osdl(dot)org>, Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us>,pgsql(at)mohawksoft(dot)com
Subject:
Re: 8.0.X and the ARC patent
Date:
2005-03-03 01:55:31
Message-ID:
15896.1109814931@sss.pgh.pa.us (view raw or flat )
Thread:
2005-02-14 23:17:48 from Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us>
2005-02-15 01:05:01 from pgsql(at)mohawksoft(dot)com
2005-02-15 01:09:24 from Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us>
2005-02-15 02:21:01 from pgsql(at)mohawksoft(dot)com
2005-02-15 07:10:12 from Thomas Hallgren <thhal(at)mailblocks(dot)com>
2005-02-15 02:56:49 from "Joshua D(dot) Drake" <jd(at)commandprompt(dot)com>
2005-02-15 03:06:02 from Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us>
2005-02-15 04:02:43 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2005-02-15 04:04:32 from Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us>
2005-02-15 05:13:45 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2005-02-19 04:15:37 from Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us>
2005-02-19 05:57:52 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2005-03-01 21:51:16 from Mark Wong <markw(at)osdl(dot)org>
2005-03-01 21:57:11 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2005-03-01 22:06:42 from Mark Wong <markw(at)osdl(dot)org>
2005-03-01 22:17:07 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2005-03-02 15:21:41 from Mark Wong <markw(at)osdl(dot)org>
2005-03-02 15:41:57 from Dave Cramer <pg(at)fastcrypt(dot)com>
2005-03-02 15:47:40 from Mark Wong <markw(at)osdl(dot)org>
2005-03-02 16:04:21 from Dave Cramer <pg(at)fastcrypt(dot)com>
2005-03-02 22:15:37 from Mark Wong <markw(at)osdl(dot)org>
2005-03-03 16:08:36 from Greg Stark <gsstark(at)mit(dot)edu>
2005-03-02 18:45:15 from Michael Adler <adler(at)pobox(dot)com>
2005-03-03 01:55:31 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2005-03-04 19:32:26 from Simon Riggs <simon(at)2ndquadrant(dot)com>
2005-03-04 19:42:27 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2005-03-05 01:10:52 from Greg Stark <gsstark(at)mit(dot)edu>
2005-03-05 18:03:17 from Simon Riggs <simon(at)2ndquadrant(dot)com>
2005-03-05 19:17:44 from Greg Stark <gsstark(at)mit(dot)edu>
2005-03-02 20:15:54 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2005-03-02 21:10:18 from Mark Wong <markw(at)osdl(dot)org>
2005-03-03 01:33:38 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2005-03-03 04:28:36 from Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us>
2005-02-21 22:54:24 from Simon Riggs <simon(at)2ndquadrant(dot)com>
2005-02-25 05:18:37 from Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us>
2005-02-25 06:47:39 from "Marc G(dot) Fournier" <scrappy(at)postgresql(dot)org>
2005-02-25 11:04:19 from Simon Riggs <simon(at)2ndquadrant(dot)com>
2005-02-25 15:36:36 from Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us>
2005-02-25 15:47:09 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2005-02-25 16:20:34 from "Matthew T(dot) O'Connor" <matthew(at)zeut(dot)net>
2005-02-25 16:24:48 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2005-02-25 16:21:06 from Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us>
2005-02-25 11:02:28 from Simon Riggs <simon(at)2ndquadrant(dot)com>
2005-02-25 11:05:44 from Andrew Dunstan <andrew(at)dunslane(dot)net>
Lists:
pgsql-hackers
Michael Adler <adler(at)pobox(dot)com> writes:
> Looking at the "Response Time Charts"
> 8.0.1/ARC
> http://www.osdl.org/projects/dbt2dev/results/dev4-010/309/rt.html
> 20050301 with 2Q patch
> http://www.osdl.org/projects/dbt2dev/results/dev4-010/313/rt.html
> It seems like the average response time has gone down, but the worse
> case ceiling has raised about 35%.
The worst cases are associated with checkpoints. I'm not sure why a
checkpoint would have a greater effect on the 2Q system than an ARC
system --- checkpoint doesn't request any new buffers so you'd think
it'd be independent. Maybe this says that the bgwriter is less
effective with 2Q, so that there are more dirty buffers remaining to
be written at the checkpoint? But why?
regards, tom lane
In response to
Responses
pgsql-hackers by date
Next :From: Bruce MomjianDate: 2005-03-03 04:28:36Subject : Re: 8.0.X and the ARC patentPrevious :From : Tom LaneDate : 2005-03-03 01:33:38Subject : Re: 8.0.X and the ARC patent
