Re: Seeking reason behind performance gain in 12 with HashAggregate

On Mon, Jan 13, 2020 at 12:44:14PM -0800, Shira Bezalel wrote:
>Hi Michael,
>
>I appreciate your question. I ran a vacuum analyze on the 9.6 table and it
>yielded no difference. Same number of buffers were read, same query plan.
>

VACUUM ANALYZE won't shrink the table - the number of buffers will be
exactly the same. You need to do VACUUM FULL, but be careful as that
acquires exclusive lock on the table.


regards

-- 
Tomas Vondra                  http://www.2ndQuadrant.com
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services