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

Stephen Frost <sfrost@snowman.net>

From: Stephen Frost <sfrost@snowman.net>
To: Merlin Moncure <mmoncure@gmail.com>
Cc: 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: 2014-04-17T19:28:20Z
Lists: pgsql-hackers
On Thursday, April 17, 2014, Merlin Moncure <mmoncure@gmail.com> wrote:
>
> yeah -- the thing is, we are already too spendy already on
> supplemental write i/o (hint bits, visible bits, freezing, etc) and
> likely not worth it to throw something else on the pile unless the
> page is already dirty; the medium term trend in storage is that read
> vs write performance is becoming increasingly asymmetric, particularly
> on the random side so it's very unlikely to balance out.
>

Guess I wasn't clear but I was thinking to read the page in, not do any
writing, and do it in a asynchronous way to the process doing the evicting.

Thanks,

Stephen

Commits

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