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

On Mon, Apr 14, 2014 at 05:45:56PM -0700, Peter Geoghegan wrote:
> On Mon, Apr 14, 2014 at 5:30 PM, Bruce Momjian <bruce@momjian.us> wrote:
> > I am glad you are looking at this.  You are right that it requires a
> > huge amount of testing, but clearly our code needs improvement in this
> > area.
> 
> Thanks.
> 
> Does anyone recall the original justification for the recommendation
> that shared_buffers never exceed 8GiB? I'd like to revisit the test
> case, if such a thing exists.

I have understood it be that the overhead of managing over 1 million
buffers is too large if you aren't accessing more than 8GB of data in a
five-minute period.  If are accessing that much, it might be possible to
have a win over 8GB.

--  Bruce Momjian  <bruce@momjian.us>        http://momjian.us EnterpriseDB
http://enterprisedb.com
 + Everyone has their own god. +