From:
Robert Haas <robertmhaas(at)gmail(dot)com>
To:
Peter Geoghegan <peter(at)2ndquadrant(dot)com>
Cc:
pgsql-hackers(at)postgresql(dot)org, Noah Misch <noah(at)leadboat(dot)com>, Simon Riggs <simon(at)2ndquadrant(dot)com>, Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>, Andres Freund <andres(at)anarazel(dot)de>, Kevin Grittner <Kevin(dot)Grittner(at)wicourts(dot)gov>, david(at)fetter(dot)org, aidan(at)highrise(dot)ca, stark(at)mit(dot)edu
Subject:
Re: 16-bit page checksums for 9.2
Date:
2012-02-23 01:39:54
Message-ID:
CA+TgmoZ2L+YfhOaGbTeiwX06t9UUP7CdAgMMEseuCB0WjsrpxQ@mail.gmail.com (view raw or flat )
Thread:
2012-01-04 13:35:19 from "Kevin Grittner" <Kevin(dot)Grittner(at)wicourts(dot)gov>
2012-01-05 15:46:22 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-01-06 18:26:14 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-01-06 19:35:01 from Andres Freund <andres(at)anarazel(dot)de>
2012-01-06 19:49:52 from Robert Haas <robertmhaas(at)gmail(dot)com>
2012-01-06 20:49:37 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-01-06 19:45:45 from Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>
2012-01-06 19:48:07 from Andres Freund <andres(at)anarazel(dot)de>
2012-01-06 19:53:38 from Robert Haas <robertmhaas(at)gmail(dot)com>
2012-01-06 20:03:49 from Andres Freund <andres(at)anarazel(dot)de>
2012-01-06 22:17:09 from Merlin Moncure <mmoncure(at)gmail(dot)com>
2012-01-06 23:48:32 from Aidan Van Dyk <aidan(at)highrise(dot)ca>
2012-01-06 23:49:04 from Aidan Van Dyk <aidan(at)highrise(dot)ca>
2012-01-06 23:48:45 from Robert Haas <robertmhaas(at)gmail(dot)com>
2012-01-06 20:47:12 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-01-06 21:44:02 from Robert Haas <robertmhaas(at)gmail(dot)com>
2012-01-07 10:14:30 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-01-07 10:26:42 from Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>
2012-01-07 10:55:19 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-01-07 11:09:53 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-01-08 14:03:46 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-01-11 22:12:31 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-01-26 20:20:39 from Noah Misch <noah(at)leadboat(dot)com>
2012-01-27 00:01:21 from Dan Scales <scales(at)vmware(dot)com>
2012-01-27 13:19:32 from Robert Haas <robertmhaas(at)gmail(dot)com>
2012-01-27 21:07:00 from Dan Scales <scales(at)vmware(dot)com>
2012-01-28 13:49:02 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-01-28 20:44:32 from Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>
2012-01-30 15:26:05 from Robert Haas <robertmhaas(at)gmail(dot)com>
2012-02-07 20:58:59 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-02-08 03:24:05 from Noah Misch <noah(at)leadboat(dot)com>
2012-02-08 11:40:34 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-02-08 14:05:02 from Noah Misch <noah(at)leadboat(dot)com>
2012-02-09 15:16:04 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-02-16 11:16:31 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-02-16 11:48:13 from Albert Cervera i Areny <albert(at)nan-tic(dot)com>
2012-02-16 13:53:48 from Robert Haas <robertmhaas(at)gmail(dot)com>
2012-02-17 22:13:20 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-02-18 11:39:13 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-02-19 16:35:48 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-02-19 19:17:13 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-02-19 22:04:06 from Robert Haas <robertmhaas(at)gmail(dot)com>
2012-02-19 23:33:19 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-02-20 00:20:07 from Robert Haas <robertmhaas(at)gmail(dot)com>
2012-02-19 23:57:20 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-02-20 00:42:01 from Robert Haas <robertmhaas(at)gmail(dot)com>
2012-02-20 09:18:44 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-02-20 13:57:08 from Robert Haas <robertmhaas(at)gmail(dot)com>
2012-02-20 17:53:55 from Josh Berkus <josh(at)agliodbs(dot)com>
2012-02-20 18:22:10 from Robert Haas <robertmhaas(at)gmail(dot)com>
2012-02-20 22:49:25 from Josh Berkus <josh(at)agliodbs(dot)com>
2012-02-20 23:02:37 from Simon Riggs <simon(at)2ndquadrant(dot)com>
2012-02-21 10:07:40 from Noah Misch <noah(at)leadboat(dot)com>
2012-02-21 14:48:42 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-02-22 13:32:04 from Robert Haas <robertmhaas(at)gmail(dot)com>
2012-02-22 23:17:53 from Peter Geoghegan <peter(at)2ndquadrant(dot)com>
2012-02-22 23:51:00 from David Fetter <david(at)fetter(dot)org>
2012-02-23 01:39:54 from Robert Haas <robertmhaas(at)gmail(dot)com>
2012-03-02 16:58:01 from Peter Geoghegan <peter(at)2ndquadrant(dot)com>
2012-02-20 23:09:01 from Bruce Momjian <bruce(at)momjian(dot)us>
2012-02-20 23:23:42 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-02-21 04:21:10 from Bruce Momjian <bruce(at)momjian(dot)us>
2012-02-22 07:06:49 from Noah Misch <noah(at)leadboat(dot)com>
2012-02-22 12:30:46 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-02-29 14:40:11 from Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>
2012-02-29 15:01:37 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-02-29 15:30:12 from Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>
2012-02-29 15:42:17 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-02-29 15:46:01 from Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>
2012-02-29 16:24:21 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-02-29 16:41:09 from Robert Haas <robertmhaas(at)gmail(dot)com>
2012-02-29 17:26:42 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-02-29 17:44:21 from Robert Haas <robertmhaas(at)gmail(dot)com>
2012-02-29 17:54:38 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-02-29 19:09:02 from Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>
2012-02-29 19:18:44 from Alvaro Herrera <alvherre(at)commandprompt(dot)com>
2012-02-29 19:29:26 from Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>
2012-02-29 20:14:14 from Alvaro Herrera <alvherre(at)commandprompt(dot)com>
2012-02-29 19:30:56 from Robert Haas <robertmhaas(at)gmail(dot)com>
2012-02-29 19:33:28 from Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>
2012-02-29 21:24:27 from Robert Haas <robertmhaas(at)gmail(dot)com>
2012-02-29 21:34:27 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2012-02-29 21:53:18 from Alvaro Herrera <alvherre(at)commandprompt(dot)com>
2012-02-29 23:38:24 from Jim Nasby <jim(at)nasby(dot)net>
2012-03-01 00:48:47 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2012-02-29 21:55:18 from Robert Haas <robertmhaas(at)gmail(dot)com>
2012-02-29 22:52:34 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2012-03-01 12:41:04 from Robert Haas <robertmhaas(at)gmail(dot)com>
2012-03-01 17:42:43 from Josh Berkus <josh(at)agliodbs(dot)com>
2012-03-01 18:28:52 from Robert Haas <robertmhaas(at)gmail(dot)com>
2012-03-01 18:40:44 from Josh Berkus <josh(at)agliodbs(dot)com>
2012-03-01 21:08:26 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2012-03-02 00:23:06 from Robert Haas <robertmhaas(at)gmail(dot)com>
2012-03-02 00:45:36 from Alvaro Herrera <alvherre(at)commandprompt(dot)com>
2012-03-02 01:32:23 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2012-03-02 01:46:18 from Robert Haas <robertmhaas(at)gmail(dot)com>
2012-03-02 02:11:31 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2012-03-02 03:52:22 from Robert Haas <robertmhaas(at)gmail(dot)com>
2012-03-02 19:32:48 from Simon Riggs <simon(at)2ndquadrant(dot)com>
2012-02-29 19:28:20 from Robert Haas <robertmhaas(at)gmail(dot)com>
2012-02-19 21:49:48 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2012-03-02 19:45:09 from Jeff Janes <jeff(dot)janes(at)gmail(dot)com>
2012-06-13 02:30:04 from Jeff Davis <pgsql(at)j-davis(dot)com>
2012-08-10 23:16:43 from Jeff Davis <pgsql(at)j-davis(dot)com>
2012-01-10 04:27:28 from Greg Smith <greg(at)2ndQuadrant(dot)com>
Lists:
pgsql-hackers
On Wed, Feb 22, 2012 at 6:17 PM, Peter Geoghegan <peter(at)2ndquadrant(dot)com> wrote:
> I decided that it would be worth benchmarking this patch.
> Specifically, I tested:
>
> master, as a basis of comparison
>
> checksum16_with_wallogged_hint_bits.v10.patch, page_checksums = 'on'
>
> checksum16_with_wallogged_hint_bits.v10.patch, page_checksums = 'off'
>
> This test was performed using pgbench-tools. At different client
> counts and scaling factors "1,10,100", performance of an update.sql
> workload was tested.
>
> This benchmark used Greg Smith's "toy" server. As he put it recently:
>
> "The main change to the 8 hyperthreaded core test server (Intel
> i7-870) for this year is bumping it from 8GB to 16GB of RAM, which
> effectively doubles the scale I can reach before things slow
> dramatically." However, while Greg used scientific Linux for his
> recent batch of performance numbers, the box was booted into Debian
> for this, which used Kernel 2.6.32, FWIW. Didn't bother with a
> separate disk for WAL.
>
> I put shared_buffers at 1GB, and checkpoint_segments at 50. I took the
> additional precaution of initdb'ing for each set, lest there be some
> kind of contamination between sets, which necessitated doing some
> additional work since I couldn't very well expect the "results"
> database to persist. Different sets of figures from different runs
> where dumped and restored, before finally being combined so that
> pgbench-tools could produce it's regular report.
>
> I have attached a config for pgbench-tools, so that others may
> recreate my work here. I also attach the most relevant image,
> comparing each test set across scaling levels. I'll make a pdf of the
> full report available if that would be useful.
Thanks for testing this. The graph obscures a bit how much percentage
change we're talking about here - could you post the raw tps numbers?
I think we also need to test the case of seq-scanning a large, unhinted table.
--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company
In response to
Responses
pgsql-hackers by date
Next :From: Daniel FarinaDate: 2012-02-23 04:17:08Subject : Re: Should we add crc32 in libpgport?Previous :From : Robert HaasDate : 2012-02-23 01:36:08Subject : Re: Displaying accumulated autovacuum cost
