Re: CPU costs of random_zipfian in pgbench

Fabien COELHO <coelho@cri.ensmp.fr>

View thread View on postgr.es

From: Fabien COELHO <coelho@cri.ensmp.fr>

To: Tomas Vondra <tomas.vondra@2ndquadrant.com>

Cc: Ants Aasma <ants.aasma@eesti.ee>, PostgreSQL Hackers <pgsql-hackers@postgresql.org>

Date: 2019-02-22T16:17:40Z

Lists: pgsql-hackers

Attachments

pgbench-zipf-doc-2.patch (text/x-diff) patch

>>> I also noticed that i is int in this function, but n is int64. That 
>>> seems like an oversight.
>
> Indeed, that is a bug!

Here is a v2 with hopefully better wording, comments and a fix for the bug 
you pointed out.

-- 
Fabien.

Commits

Remove now-unnecessary thread pointer arguments in pgbench.
- d8c0bd9fefa9 12.0 landed
Restrict pgbench's zipfian parameter to ensure good performance.
- 26a76cb64072 12.0 landed