On Tue, 2009-05-12 at 11:30 -0700, Scott Carey wrote:
> the fact is there is no evidence that a connection pooler will fix the
> scalability from 16 > 32 cores.
There has been much analysis over a number of years of the effects of
the ProcArrayLock, specifically the O(N^2) effect of increasing numbers
of connections on GetSnapshotData(). Most discussion has been on
-hackers, not -perform.
--
Simon Riggs www.2ndQuadrant.com
PostgreSQL Training, Services and Support
