Horton

Re: simple patch for discussion

David G. Johnston <david.g.johnston@gmail.com>

View thread
From: "David G. Johnston" <david.g.johnston@gmail.com>
To: David Rowley <dgrowleyml@gmail.com>
Cc: Andres Freund <andres@anarazel.de>, Greg Hennessy <greg.hennessy@gmail.com>, "pgsql-hackers@postgresql.org" <pgsql-hackers@postgresql.org>
Date: 2025-07-18T04:23:57Z
Lists: pgsql-hackers 
On Thursday, July 17, 2025, David G. Johnston <david.g.johnston@gmail.com>
wrote:

> On Thursday, July 17, 2025, David Rowley <dgrowleyml@gmail.com> wrote:
>
>>
>> I find it hard to imagine that there'd be many people
>>
> around that would benefit from a global
>> "always_use_this_number_of_parallel_workers" GUC.
>>
>
> Just to clarify, the global value would be -1 usually.  It would be
> another crappy planner hint for important queries to be assigned saying
> “make this query go fast even if it doesn’t play nice with others”.
>

Framing this differently, how about a patch that lets extension authors
choose to implement alternative formulas or even provide GUC-driven
constants into the planner at the existing spot instead of having to choose
a best algorithm.  IOW, what would it take to make the proposed patch an
extension that a DBA could choose to install and override the current log3
algorithm?

David J.