Re: [HACKERS] advanced partition matching algorithm for partition-wise join

Robert Haas <robertmhaas@gmail.com>

From: Robert Haas <robertmhaas@gmail.com>

To: Dmitry Dolgov <9erthalion6@gmail.com>

Cc: Amit Langote <Langote_Amit_f8@lab.ntt.co.jp>, Ashutosh Bapat <ashutosh.bapat@enterprisedb.com>, Antonin Houska <ah@cybertec.at>, PostgreSQL-development <pgsql-hackers@postgresql.org>

Date: 2018-07-16T15:01:51Z

Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

Suppress unused-variable warning.
- 401418ca6a68 13.0 landed
Allow partitionwise joins in more cases.
- c8434d64ce03 13.0 landed
Avoid crash in partitionwise join planning under GEQO.
- 7ad6498fd5a6 12.0 cited
- d70c147fa217 11.3 cited
Disable support for partitionwise joins in problematic cases.
- 7cfdc77023ad 12.0 cited
Add plan_cache_mode setting
- f7cb2842bf47 12.0 cited
Add test for partitionwise join involving default partition.
- 4513d3a4be0b 12.0 cited
Revise API for partition_rbound_cmp/partition_rbound_datum_cmp.
- b0229235564f 11.0 landed

On Sun, Jul 15, 2018 at 1:43 PM, Dmitry Dolgov <9erthalion6@gmail.com> wrote:
> Partitions: test11 FOR VALUES FROM (0) TO (100),
>             test12 FOR VALUES FROM (100) TO (200),
>             test13 FOR VALUES FROM (200) TO (300)
>
> Partitions: test21 FOR VALUES FROM (10) TO (110),
>             test22 FOR VALUES FROM (110) TO (210),
>             test23 FOR VALUES FROM (210) TO (310)
>
> I'm confused, since there is only one-to-one mapping of partitions.

No, test21 would have to be joined to both test11 and test12, since
either could contain matching rows.  Also, test22 would have to be
joined to both test12 and test13.

-- 
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company