"Jatinder Sangha" <js@xxxxxxxxxxxxx> wrote: > I've already converted all of my SQL to use "distinct on (...)" > and this is now always faster using the hash-aggregates than when > using sorting. The queries now only use sorting if the hashing > would take up too much memory. It's great that you have a solution to your immediate problem, but if the active portion of your database is really as fully cached as your problem case indicates, you should probably still tweak the costing factors. Doing so will help the optimizer pick good plans for any arbitrary query you choose to run. If the caching was unusual, and was just showing up because were repeatedly running that one test case, then never mind. :-) -Kevin -- Sent via pgsql-performance mailing list (pgsql-performance@xxxxxxxxxxxxxx) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-performance
