Horton

Re: Clock sweep not caching enough B-Tree leaf pages?

Tom Lane <tgl@sss.pgh.pa.us>

View thread View on postgr.es
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Stephen Frost <sfrost@snowman.net>
Cc: Merlin Moncure <mmoncure@gmail.com>, Andres Freund <andres@2ndquadrant.com>, Heikki Linnakangas <hlinnakangas@vmware.com>, Greg Stark <stark@mit.edu>, Robert Haas <robertmhaas@gmail.com>, Peter Geoghegan <pg@heroku.com>, Pg Hackers <pgsql-hackers@postgresql.org>
Date: 2014-04-17T19:09:43Z
Lists: pgsql-hackers 
Stephen Frost <sfrost@snowman.net> writes:
> I wonder if it would help to actually tell the OS to read in buffers
> that we're *evicting*...  On the general notion that if the OS already
> has them buffered then it's almost a no-op, and if it doesn't and it's
> actually a 'hot' buffer that we're gonna need again shortly, the OS will
> have it.

But if it's actually gone cold, you're just forcing unnecessary read I/O,
not to mention possibly causing something slightly warmer to be lost from
kernel cache.

			regards, tom lane

Commits

  1. Replace the BufMgrLock with separate locks on the lookup hashtable and