|From:||Gavin Flower <GavinFlower(at)archidevsys(dot)co(dot)nz>|
|To:||Amit Kapila <amit(dot)kapila16(at)gmail(dot)com>, Robert Haas <robertmhaas(at)gmail(dot)com>|
|Cc:||Michael Banck <michael(dot)banck(at)credativ(dot)de>, Anastasia Lubennikova <a(dot)lubennikova(at)postgrespro(dot)ru>, Anastasia Lubennikova <lubennikovaav(at)gmail(dot)com>, pgsql-hackers <pgsql-hackers(at)postgresql(dot)org>, Rahila Syed <rahilasyed(dot)90(at)gmail(dot)com>|
|Subject:||Re: Parallel Index Scans|
|Views:||Raw Message | Whole Thread | Download mbox|
On 07/03/17 02:46, Amit Kapila wrote:
> On Mon, Mar 6, 2017 at 6:49 PM, Robert Haas <robertmhaas(at)gmail(dot)com> wrote:
>> On Mon, Mar 6, 2017 at 6:33 AM, Amit Kapila <amit(dot)kapila16(at)gmail(dot)com> wrote:
>> I was going to do it after index and index-only scans and parallel
>> bitmap heap scan were committed, but I've been holding off on
>> committing parallel bitmap heap scan waiting for Andres to fix the
>> simplehash regressions, so maybe I should just go do an update now and
>> another one later once that goes in.
> As you wish, but one point to note is that as of now parallelism for
> index scans can be influenced by table level parameter
> parallel_workers. It sounds slightly awkward, but if we want to keep
> that way, then maybe we can update the docs to indicate the same.
> Another option is to have a separate parameter for index scans.
My immediate gut feeling was to have separate parameters.
On thinking about it, I think they serve different use cases. I don't
think of workers when I think of Index scans, and I suspect I'd find
more reasons to keep them separate if I looked deeper.
|Next Message||Tomas Vondra||2017-03-06 18:49:56||Re: PATCH: two slab-like memory allocators|
|Previous Message||Peter Eisentraut||2017-03-06 18:45:21||Re: Other formats in pset like markdown, rst, mediawiki|