Horton

Re: Why is DEFAULT_FDW_TUPLE_COST so insanely low?

David Rowley <dgrowleyml@gmail.com>

View thread View on postgr.es
From: David Rowley <dgrowleyml@gmail.com>
To: Richard Guo <guofenglinux@gmail.com>
Cc: Michael Paquier <michael@paquier.xyz>, Bruce Momjian <bruce@momjian.us>, PostgreSQL Developers <pgsql-hackers@lists.postgresql.org>
Date: 2023-11-02T23:38:26Z
Lists: pgsql-hackers 
On Fri, 3 Nov 2023 at 01:02, Richard Guo <guofenglinux@gmail.com> wrote:
> It seems that the test is still not stable on 32-bit machines even after
> 4b14e18714.  I see the following plan diff on cfbot [1].

I recreated that locally this time.  Seems there's still flexibility
to push or not push down the sort and the costs of each are close
enough that it differs between 32 and 64-bit.

The fix I just pushed removes the flexibility for doing a local sort
by turning off enable_sort.

Thanks for the report.

David

Commits

  1. Stabilize postgres_fdw tests on 32-bit machines

  2. Attempt to stabilize postgres_fdw tests

  3. Increase DEFAULT_FDW_TUPLE_COST from 0.01 to 0.2

  4. Add postgres_fdw contrib module.