On 05/07/2014 11:52 AM, Peter Geoghegan wrote:
> On Wed, May 7, 2014 at 11:40 AM, Josh Berkus <josh@agliodbs.com> wrote:
>> So, as one of several people who put literally hundreds of hours into
>> the original benchmarking which established the sizing recommendations
>> for shared_buffers (and other settings), I find the phrase "folk wisdom"
>> personally offensive. So, can we stop with this?
>
> I have also put a lot of time into benchmarking. No personal offence
> was intended, and I'm glad that we have some advice to give to users,
> but the fact of the matter is that current *official* recommendations
> are very vague.
Well, they should be vague; the only hard data we have is rather
out-of-date (I think 8.2 was our last set of tests). If we gave users
specific, detailed recommendations, we'd be misleading them.
For that matter, our advice on shared_buffers ... and our design for it
... is going to need to change radically soon, since Linux is getting an
ARC with a frequency cache as well as a recency cache, and FreeBSD and
OpenSolaris already have them.
FWIW, if someone could fund me for a month, I'd be happy to create a
benchmarking setup where we could test these kinds of things; I have
pretty clear ideas how to build one. I imagine some of our other
consultants could make the same offer. However, it's too much work for
anyone to get done "in their spare time".
--
Josh Berkus
PostgreSQL Experts Inc.
http://pgexperts.com