Re: Mysteriously varying index scan costs

[Pavan Deolasee <pavan.deolasee@xxxxxxxxx>]

On Mon, Sep 24, 2018 at 2:20 PM, Laurenz Albe <laurenz.albe@xxxxxxxxxxx> wrote:

But how can it be that the first run has to touch 74917 blocks,
while whe second run only needs to touch 1185?

The first index scan may have killed lots of index tuples.

Thanks,

Pavan 

--

 Pavan Deolasee                   http://www.2ndQuadrant.com/
 PostgreSQL Development, 24x7 Support, Training & Services