Re: [WIP] speeding up GIN build with parallel workers - Mailing list pgsql-hackers

From Constantin S. Pan
Subject Re: [WIP] speeding up GIN build with parallel workers
Date
Msg-id 20160316125539.01cd4fd9@ppg
Whole thread Raw
In response to Re: [WIP] speeding up GIN build with parallel workers  (Peter Geoghegan <pg@heroku.com>)
List pgsql-hackers
On Wed, 16 Mar 2016 02:43:47 -0700
Peter Geoghegan <pg@heroku.com> wrote:

> On Wed, Mar 16, 2016 at 2:25 AM, Constantin S. Pan <kvapen@gmail.com>
> wrote:
> > The backend just waits for the results from the workers and merges
> > them (in case wnum > 0). So the 1-worker configuration should never
> > be used, because it is as sequential as the 0-worker, but adds data
> > transfer.
> 
> This is why I wanted an easy way of atomically guaranteeing some
> number of workers (typically 2), or not using parallelism at all. I
> think the parallel worker API should offer a simple way to do that in
> cases like this, where having only 1 worker is never going to win.

Well, we can check the number of workers actually launched and revert
back to single backend way when there is less than 2 workers. Let me
code that in.

Regards,

Constantin S. Pan
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company



pgsql-hackers by date:

Previous
From: Peter Geoghegan
Date:
Subject: Re: [WIP] speeding up GIN build with parallel workers
Next
From: Andreas Karlsson
Date:
Subject: Re: syslog configurable line splitting behavior