On Fri, 2 Jul 2021 at 10:24, David Rowley <dgrowleyml@gmail.com> wrote:
>
> I ran this again with a few different worker counts after tuning a few
> memory settings so there was no spilling to disk and so everything was
> in RAM. Mostly so I could get consistent results.
>
> Here's the results. Average over 3 runs on each:
>
> Workers Master Patched Percent
> 8 11094.1 11084.9 100.08%
> 16 8711.4 8562.6 101.74%
> 32 6961.4 6726.3 103.50%
> 64 6137.4 5854.8 104.83%
> 128 6090.3 5747.4 105.96%
>
Thanks for testing again. Those are nice looking results, and are much
more in line with what I was seeing.
> So the gains are much less at lower worker counts. I think this is
> because most of the gains are in the serial part of the plan and with
> higher worker counts that part of the plan is relatively much bigger.
>
> So likely performance isn't too critical here, but it is something to
> keep in mind.
>
Yes, agreed. I suspect there's not much more that can be shaved off
this particular piece of code now though. Here's an update with the
last set of changes discussed.
Regards,
Dean
