From: Tom Lane <tgl@sss.pgh.pa.us>
To: Jeff Janes <jeff.janes@gmail.com>
cc: "pgsql-performa." <pgsql-performance@postgresql.org>
Subject: Re: How to set parallel_tuple_cost
In-reply-to: 
 <CAMkU=1ziWTO_OBmFYkwhdtpc_JrM2Eroip=8vS3CNtgJs6KcoQ@mail.gmail.com>
References: 
 <CAMkU=1ziWTO_OBmFYkwhdtpc_JrM2Eroip=8vS3CNtgJs6KcoQ@mail.gmail.com>
Comments: In-reply-to Jeff Janes <jeff.janes@gmail.com>
	message dated "Fri, 20 Dec 2019 13:03:29 -0500"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <11001.1576868337.1@sss.pgh.pa.us>
Date: Fri, 20 Dec 2019 13:58:57 -0500
Message-ID: <11002.1576868337@sss.pgh.pa.us>
Precedence: bulk

Jeff Janes <jeff.janes@gmail.com> writes:
> The docs for parallel_tuple_cost are quite terse, as the reference section
> of the docs usually are:
> "Sets the planner's estimate of the cost of transferring one tuple from a
> parallel worker process to another process. The default is 0.1."

> If you were take the doc description literally, then the default value
> seems much too high, as it doesn't take 10x the (default) cpu_tuple_cost to
> transfer a tuple up from a parallel worker.

Really?  If anything, I'd have thought it might be worse than 10x.
Cross-process communication isn't cheap, at least not according to
my instincts.

> On the other hand, you
> probably don't want a query which consumes 8x the CPU resources just to
> finish only 5% faster (on an otherwise idle server with 8 CPUs).  Maybe
> this Amdahl factor is what inspired the high default value?

I think the large value of parallel_setup_cost is what's meant to
discourage that scenario.

			regards, tom lane