Robert Haas <robertmhaas@xxxxxxxxx> writes: > On Thu, Sep 24, 2009 at 8:25 PM, Josh Berkus <josh@xxxxxxxxxxxx> wrote: >> You can't "turn off" sequential scans. You can only make the planner >> less likely to choose them. But if there's no way to get the data you >> need other than a seqscan, it's still going to do one. > And that's not a bad thing. For a very small table, it's often the > fastest method. Probably more to the point: if the query involves fetching the whole table, it's *always* the fastest method. (Except maybe if you want the results sorted, and often it's the fastest way even so.) Indexes are not a panacea. regards, tom lane -- Sent via pgsql-performance mailing list (pgsql-performance@xxxxxxxxxxxxxx) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-performance
