From:
Mark Kirkwood <mark(dot)kirkwood(at)catalyst(dot)net(dot)nz>
To:
Robert Haas <robertmhaas(at)gmail(dot)com>
Cc:
Simon Riggs <simon(at)2ndquadrant(dot)com>, David Fetter <david(at)fetter(dot)org>,
Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>,
Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>,
Erik Rijkers <er(at)xs4all(dot)nl>,
pgsql-hackers(at)postgresql(dot)org
Subject:
Re: testing HS/SR - 1 vs 2 performance
Date:
2010-04-21 03:09:43
Message-ID:
4BCE6C77.6060909@catalyst.net.nz (view raw or flat )
Thread:
2010-04-12 14:00:51 from "Erik Rijkers" <er(at)xs4all(dot)nl>
2010-04-13 05:44:19 from Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>
2010-04-13 18:09:46 from Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>
2010-04-14 08:24:45 from Dimitri Fontaine <dfontaine(at)hi-media(dot)com>
2010-04-14 12:08:13 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-16 08:29:54 from Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>
2010-04-16 09:10:01 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-16 11:47:42 from Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>
2010-04-16 12:00:22 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-16 21:00:09 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-16 14:39:52 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2010-04-16 14:52:49 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-16 15:10:45 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2010-04-17 14:16:55 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-17 15:13:46 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2010-04-17 15:25:46 from Robert Haas <robertmhaas(at)gmail(dot)com>
2010-04-17 19:20:00 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-17 19:45:47 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2010-04-17 20:14:33 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-17 20:48:21 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2010-04-17 22:46:06 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-17 22:52:26 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2010-04-18 07:24:36 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-18 11:01:05 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-18 20:16:11 from David Fetter <david(at)fetter(dot)org>
2010-04-18 20:22:21 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-18 20:23:21 from David Fetter <david(at)fetter(dot)org>
2010-04-20 11:50:23 from Robert Haas <robertmhaas(at)gmail(dot)com>
2010-04-21 03:09:43 from Mark Kirkwood <mark(dot)kirkwood(at)catalyst(dot)net(dot)nz>
2010-04-21 03:46:06 from Robert Haas <robertmhaas(at)gmail(dot)com>
2010-04-21 06:39:16 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-21 12:20:32 from Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>
2010-04-21 13:31:28 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-21 14:11:00 from Robert Haas <robertmhaas(at)gmail(dot)com>
2010-04-21 12:27:12 from Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>
2010-04-21 13:37:46 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-21 13:51:00 from Robert Haas <robertmhaas(at)gmail(dot)com>
2010-04-21 14:12:43 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-21 14:22:50 from marcin mank <marcin(dot)mank(at)gmail(dot)com>
2010-04-21 14:49:06 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-21 15:13:29 from Florian Pflug <fgp(at)phlo(dot)org>
2010-04-21 14:53:05 from Robert Haas <robertmhaas(at)gmail(dot)com>
2010-04-22 05:57:43 from Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>
2010-04-22 06:56:39 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-22 18:39:14 from "Erik Rijkers" <er(at)xs4all(dot)nl>
2010-04-22 21:41:29 from Greg Smith <greg(at)2ndquadrant(dot)com>
2010-04-22 21:54:55 from Mark Kirkwood <mark(dot)kirkwood(at)catalyst(dot)net(dot)nz>
2010-04-22 22:32:44 from "Erik Rijkers" <er(at)xs4all(dot)nl>
2010-04-22 22:45:47 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-23 15:14:41 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-23 15:32:19 from Robert Haas <robertmhaas(at)gmail(dot)com>
2010-04-23 22:39:48 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-23 23:07:15 from Robert Haas <robertmhaas(at)gmail(dot)com>
2010-04-24 09:17:44 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-25 10:53:04 from Robert Haas <robertmhaas(at)gmail(dot)com>
2010-04-25 12:50:00 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-25 15:37:04 from Robert Haas <robertmhaas(at)gmail(dot)com>
2010-04-25 15:50:17 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2010-04-25 16:46:38 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-25 16:51:58 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2010-04-25 17:00:58 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-25 17:33:24 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2010-04-25 17:45:02 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-25 17:51:00 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2010-04-27 02:35:22 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-27 02:37:18 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2010-04-27 17:52:50 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2010-04-27 18:13:33 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-27 18:53:16 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2010-04-27 19:29:13 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-27 20:18:53 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2010-04-27 20:44:32 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-27 21:24:48 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2010-04-27 21:45:53 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-27 22:08:29 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2010-04-27 22:45:54 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-28 00:13:09 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2010-04-28 09:37:38 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-23 23:17:22 from "Erik Rijkers" <er(at)xs4all(dot)nl>
2010-04-25 18:25:16 from "Erik Rijkers" <er(at)xs4all(dot)nl>
2010-04-25 18:55:02 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2010-04-25 21:52:19 from "Erik Rijkers" <er(at)xs4all(dot)nl>
2010-04-26 07:43:09 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-26 15:04:09 from "Erik Rijkers" <er(at)xs4all(dot)nl>
2010-04-25 19:07:28 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-05-04 16:10:53 from "Erik Rijkers" <er(at)xs4all(dot)nl>
2010-05-04 16:19:47 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-05-04 16:33:26 from "Erik Rijkers" <er(at)xs4all(dot)nl>
2010-05-04 18:26:58 from Greg Smith <greg(at)2ndquadrant(dot)com>
2010-05-04 19:40:12 from "Erik Rijkers" <er(at)xs4all(dot)nl>
2010-05-07 12:04:02 from Greg Smith <greg(at)2ndquadrant(dot)com>
2010-05-04 19:34:53 from Stefan Kaltenbrunner <stefan(at)kaltenbrunner(dot)cc>
2010-05-04 20:48:07 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-26 06:52:51 from Fujii Masao <masao(dot)fujii(at)gmail(dot)com>
2010-04-26 15:01:41 from "Erik Rijkers" <er(at)xs4all(dot)nl>
2010-04-23 00:44:11 from Mark Kirkwood <mark(dot)kirkwood(at)catalyst(dot)net(dot)nz>
2010-04-25 16:43:57 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2010-04-25 16:54:59 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-25 23:18:48 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2010-04-26 07:21:35 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2010-04-23 17:16:09 from Marko Kreen <markokr(at)gmail(dot)com>
2010-04-23 17:38:48 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2010-04-23 18:22:55 from Marko Kreen <markokr(at)gmail(dot)com>
Lists:
pgsql-hackers
Robert Haas wrote:
>
> So, does anyone have a few cycles to test this out? We are down to
> handful of remaining open items, so getting this tested and committed
> sooner = beta sooner.
>
>
>
I did some testing of this patch (v2). Unfortunately I don't have access
to hardware capable of doing tests at the same scale as Erik used.
However I was still able to show a consistent difference (I think)
between standby performance with and without the patch applied.
Setup:
host: 2.7 Ghz dual core amd64 with 4G ram and 1 sata drive,
code: cvs head from 2010-04-14.
pgbench: scale=100, 4 clients, 10000 (select) transactions each.
Results:
Master performance (with and without patch applied ):
tps = 10903.612340 - 14070.109951 (including connections establishing)
Standby performance without patch (:
tps = 8288.119913 - 9722.245178 (including connections establishing)
Standby performance with patch applied:
tps = 11592.922637 - 14065.553214 (including connections establishing)
I performed 8 runs of each, and results would start at the low range and
climb up to the high one, where they would stabilize. In between runs I
cleared the os buffer cache and (partially) reloaded it by selecting
counts from the pgbench tables (i.e I was trying to ensure each run had
the same or similar os cache contents).
Overall looks like the patch gets standby read only performance close to
the master - at least in the case where there are minimal master
transactions being tracked by the standby (I had to leave the master
idle whilst running the standby case, as they shared the machine). Hope
this info is useful.
regards
Mark
In response to
Responses
pgsql-hackers by date
Next :From: Tom LaneDate: 2010-04-21 03:33:19Subject : Re: Should database = all in pg_hba.conf match a replication connection? Previous :From : Fujii MasaoDate : 2010-04-21 01:52:51Subject : Re: Should database = all in pg_hba.conf match a replication connection?
