On Tue, Feb 03, 2026 at 11:33:20AM -0500, Tom Lane wrote:
> Remind me again why anyone would choose to run Postgres on this
> platform? Why are we moving mountains to make it possible?
Based on the numbers you are posting for the environment tested, I
cannot think of a sane reason that somebody would do that. That
points to the same feeling that I always have when reading this
thread: as far as I understand it is not really possible to have an
access to hardware that one could use to test any of this stuff or
even write patches to support it. If I/O is bad in any environments
where a patch is run, the time investment is moot compared to other
nix or even WIN32 environments where a full check-world usually takes
no more than 10 minutes for one branch, even on a laptop.
Saying that, I'm really glad for the report that has led to the
discovery fixed in d46aa32ea5ce. Opportunistic improvements not
specific to AIX are also good to know about: they show long-term
benefits everywhere.
--
Michael
