Tom Lane writes:

> Merlin Moncure <> writes:
>> Aside from the functional issues, could your changes result in
>> performance regressions?
> It's a little bit harder to gauge the impact on planner speed.  The
> transitive closure calculation could be expensive in a query with many
> lateral references, but that doesn't seem likely to be common; and anyway
> we'll buy back some of that cost due to simpler tests later.  I'm
> optimistic that we'll come out ahead in HEAD/9.5 after the removal
> of LateralJoinInfo setup.  It might be roughly a wash in the back
> branches.

On the empirical side: I see a speedup of 0.4% in testing speed with the
patch applied.  It could very well be me venting the room one additional
time during the second session, resulting in the CPUs spending more time
in their opportunistic frequency range or something.


smith=# select extract('day' from t),
               avg(generated/extract(epoch from updated - t)) as tps
        from instance natural join stat
        where generated > 1000000
              and hostname not in('dwagon','slugbug')
                           -- these do stuff beside sqlsmith
              and t > now() - interval '3 days' group by 1 order by 1;
 date_part |       tps
         8 |  55.494181110456
         9 | 55.6902316869404

Sent via pgsql-hackers mailing list (
To make changes to your subscription:

Reply via email to