Re: Partitioning Option?

Tom Lane wrote:
> "Tomeh, Husam" <htomeh(at)firstam(dot)com> writes:
>
>>(I was referring to object partitioning. For instance, if I have a huge
>>table with US counties as my partition key, I could create partitions
>>within the same table based on the partition key (a US county for
>>example). When querying, the engine will access the partition instead of
>>the whole table to get the result set. This is provided in Oracle DB EE.
>>So, I was wondering whether I can do similar thing in PostgreSQL since
>>we're exploring PostgreSQL)
>
>
> You can build it out of spare parts: either a view over a UNION ALL of
> component tables, or a parent table with a bunch of inheritance
> children, either way with rules to redirect insertions into the
> right subtable. (With the inheritance way you could instead use
> a trigger for that, which'd likely be more flexible.)

Tom, I did a post on performance about my attempt to do an horizontal partition,
in a 7.4.x engine, but it seems the planner refuse to optimize it,
look at this for example:

CREATE TABLE user_logs_2003_h () inherits (user_logs);
CREATE TABLE user_logs_2002_h () inherits (user_logs);

I defined on these tables the index already defined on user_logs.

And this is the result:

empdb=# explain analyze select * from user_logs where id_user = sp_id_user('kalman');
QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Result (cost=0.00..426.33 rows=335 width=67) (actual time=20.891..129.218 rows=98 loops=1)
-> Append (cost=0.00..426.33 rows=335 width=67) (actual time=20.871..128.643 rows=98 loops=1)
-> Index Scan using idx_user_user_logs on user_logs (cost=0.00..133.11 rows=66 width=67) (actual time=20.864..44.594 rows=3 loops=1)
Index Cond: (id_user = 4185)
-> Index Scan using idx_user_user_logs_2003_h on user_logs_2003_h user_logs (cost=0.00..204.39 rows=189 width=67) (actual time=1.507..83.662 rows=95 loops=1)
Index Cond: (id_user = 4185)
-> Index Scan using idx_user_user_logs_2002_h on user_logs_2002_h user_logs (cost=0.00..88.83 rows=80 width=67) (actual time=0.206..0.206 rows=0 loops=1)
Index Cond: (id_user = 4185)
Total runtime: 129.500 ms
(9 rows)

that is good, but now look what happen in a view like this one ( where I join the view above ):

create view to_delete AS
SELECT v.login,
u.*
from user_login v,
user_logs u
where v.id_user = u.id_user;

empdb=# explain analyze select * from to_delete where login = 'kalman';
QUERY PLAN
------------------------------------------------------------------------------------------------------------------------------------------------
Hash Join (cost=4.01..65421.05 rows=143 width=79) (actual time=1479.738..37121.511 rows=98 loops=1)
Hash Cond: ("outer".id_user = "inner".id_user)
-> Append (cost=0.00..50793.17 rows=2924633 width=67) (actual time=21.391..33987.363 rows=2927428 loops=1)
-> Seq Scan on user_logs u (cost=0.00..7195.22 rows=411244 width=67) (actual time=21.385..5641.307 rows=414039 loops=1)
-> Seq Scan on user_logs_2003_h u (cost=0.00..34833.95 rows=2008190 width=67) (actual time=0.024..18031.218 rows=2008190 loops=1)
-> Seq Scan on user_logs_2002_h u (cost=0.00..8764.00 rows=505199 width=67) (actual time=0.005..5733.554 rows=505199 loops=1)
-> Hash (cost=4.00..4.00 rows=2 width=16) (actual time=0.195..0.195 rows=0 loops=1)
-> Index Scan using user_login_login_key on user_login v (cost=0.00..4.00 rows=2 width=16) (actual time=0.155..0.161 rows=1 loops=1)
Index Cond: ((login)::text = 'kalman'::text)
Total runtime: 37122.069 ms
(10 rows)

I did a similar attempt with UNION ALL but the result is the same.

Regards
Gaetano Mendola