Horton

Re: our buffer replacement strategy is kind of lame

daveg <daveg@sonic.net>

View thread
From: daveg <daveg@sonic.net>
To: Simon Riggs <simon@2ndQuadrant.com>
Cc: Robert Haas <robertmhaas@gmail.com>, pgsql-hackers <pgsql-hackers@postgresql.org>
Date: 2011-08-12T22:42:46Z
Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. Split work of bgwriter between 2 processes: bgwriter and checkpointer.

On Fri, Aug 12, 2011 at 01:28:49PM +0100, Simon Riggs wrote:
> I think there are reasonable arguments to make
> 
> * prefer_cache = off (default) | on a table level storage parameter,
> =on will disable the use of BufferAccessStrategy
> 
> * make cache_spoil_threshold a parameter, with default 0.25
> 
> Considering the world of very large RAMs in which we now live, some
> control of the above makes sense.

As long as we are discussion cache settings for tables, I have a client
who would like to be able to lock specific tables and indexes into cache
as they have strict response time requirements for particular queries.
At the moment they are running postgres with a tablespace on ramfs and
taking frequent backups, but this is not optimal.

-dg

-- 
David Gould       daveg@sonic.net      510 536 1443    510 282 0869
If simplicity worked, the world would be overrun with insects.