| From: | Robert Haas <robertmhaas(at)gmail(dot)com> |
|---|---|
| To: | Peter Geoghegan <pg(at)bowt(dot)ie> |
| Cc: | Thomas Munro <thomas(dot)munro(at)enterprisedb(dot)com>, Rushabh Lathia <rushabh(dot)lathia(at)gmail(dot)com>, Heikki Linnakangas <hlinnaka(at)iki(dot)fi>, Pg Hackers <pgsql-hackers(at)postgresql(dot)org>, Corey Huinker <corey(dot)huinker(at)gmail(dot)com> |
| Subject: | Re: [HACKERS] Parallel tuplesort (for parallel B-Tree index creation) |
| Date: | 2018-01-18 14:21:46 |
| Message-ID: | CA+TgmoZerO==zkKn7qqZeUW=mMpc+ZRiDuTXtj8pgzMOyrag3A@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Wed, Jan 17, 2018 at 10:22 PM, Peter Geoghegan <pg(at)bowt(dot)ie> wrote:
> As I said in a prior e-mail, even parallel query's use of
> parallel_leader_participation is consistent with what I propose here,
> practically speaking, because a partial path without leader
> participation will always lose to a serial sequential scan path in
> practice.
Amit's reply to this part drew my attention to it. I think this is
entirely false. Consider an aggregate that doesn't support partial
aggregation, and a plan that looks like this:
Aggregate
-> Gather
-> Parallel Seq Scan
Filter: something fairly selective
It is quite possible for this to be superior to a non-parallel plan
even with only 1 worker and no parallel leader participation. The
worker can evaluate the filter qual, and the leader can evaluate the
aggregate. If the CPU costs of doing those computations are high
enough to outweigh the costs of shuffling tuples between backends, we
win.
--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Andrey Borodin | 2018-01-18 14:41:34 | Re: [HACKERS] WIP: Covering + unique indexes. |
| Previous Message | Amit Kapila | 2018-01-18 14:14:55 | Re: [HACKERS] Parallel tuplesort (for parallel B-Tree index creation) |