On Sun, Nov 07, 2004 at 09:29:30 +0000,
Jerry III <jerryiii@hotmail.com> wrote:
> But if you do build an index over "id" then pgsql would only have to do a
> sequential scan on that index, which might be a lot faster if your table
> contains a lot of other data, won't it?
A full table index scan will be slower than a sequential scan; typically by
a lot. In the old days a sort step would have been needed and that would have
slowed things down. Now a method using hashing is available that will
work unless there is an extremely large number of unique values for "id".