Re: Slow counting still true?

On Mon, Sep 17, 2012 at 1:07 AM, Thomas Guettler  wrote:
> Release 9.2 should increase count(*) performance. Is this wiki page still
> valid?

Even with index only scans, it still has to walk through every tuple,
even if they are only index tuples not table tuples.  So nothing has
fundamentally changed.  So it is still true that "PostgreSQL must walk
through all rows, in some sense".

I guess this sentence is partly out of date: "PostgreSQL will still
need to read the resulting rows to verify that they exist; other
database systems may only need to reference the index in this
situation."  Maybe just deleting it would be appropriate.

Cheers,

Jeff