Re: [HACKERS] Autovacuum Improvements
Martijn van Oosterhout <kleptog@svana.org>
From: Martijn van Oosterhout <kleptog@svana.org>
To: Gregory Stark <stark@enterprisedb.com>
Cc: Bruce Momjian <bruce@momjian.us>, Heikki Linnakangas <heikki@enterprisedb.com>, Russell Smith <mr-russ@pws.com.au>, Darcy Buskermolen <darcyb@commandprompt.com>, Simon Riggs <simon@2ndquadrant.com>, Alvaro Herrera <alvherre@commandprompt.com>, "Matthew T. O'Connor" <matthew@zeut.net>, Pavan Deolasee <pavan@enterprisedb.com>, Christopher Browne <cbbrowne@acm.org>, pgsql-general@postgresql.org, pgsql-hackers@postgresql.org
Date: 2007-01-23T07:27:28Z
Lists: pgsql-hackers
On Mon, Jan 22, 2007 at 05:51:53PM +0000, Gregory Stark wrote: > Actually no. A while back I did experiments to see how fast reading a file > sequentially was compared to reading the same file sequentially but skipping > x% of the blocks randomly. The results were surprising (to me) and depressing. > The breakeven point was about 7%. I asusume this means you were reading 7% of the blocks, not skipping 7% of the blocks when you broke even? I presume by break-even you mean it took just as long, time-wise. But did it have the same effect on system load? If reading only 7% of the blocks allows the drive to complete other requests more quickly then it's beneficial, even if the vacuum takes longer. This may be a silly thought, I'm not sure how drives handle multiple requests... Have a nice day, -- Martijn van Oosterhout <kleptog@svana.org> http://svana.org/kleptog/ > From each according to his ability. To each according to his ability to litigate.