On Fri, May 06, 2011 at 12:45:23PM -0600, Scott Ribe wrote: > I need to optimize queries that deal with some aggregates regarding resource availability. My specific problem is, I think, very closely analogous to select count(*)... where... If the WHERE clause is fairly selective and indexed, that should be fast. Not as fast as estimates based on trigger-written values in another table, of course, but reasonably fast. So the first order of business is usually to find or create indexes that will make SELECT on the same criteria fast. It's only unqualified "SELECT count(*)" that is slow. Generally, the system table is good enough for that, I find. (Someone: "How long will this take?" Me: "There are about 400 million rows to go through." Even if you're off by 50 million at that point, it doesn't matter.) A -- Andrew Sullivan ajs@xxxxxxxxxxxxxxx -- Sent via pgsql-general mailing list (pgsql-general@xxxxxxxxxxxxxx) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-general
