Performance select distinct sur requête simple
Bonjour,
J'ai un problème avec la requête suivante :
Code:
1 2
|
select distinct(year) from selling where type in ('A', 'C') and country like ('%FR%'); |
Le résultat est retournée après 30s (4 lignes).
Un select * est instantané.
La table contient 1,2 millions de lignes.
J'ai fait un explain mais le résultat me parle pas vraiment:
Code:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
|
Plan hash value: 188663770
-----------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
-----------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 5 | 90 | 68344 (1)| 00:13:41 |
| 1 | HASH UNIQUE | | 5 | 90 | 68344 (1)| 00:13:41 |
|* 2 | TABLE ACCESS FULL| SELLING| 1252 | 22536 | 68343 (1)| 00:13:41 |
-----------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
2 - filter("COUNTRY" IS NOT NULL AND "COUNTRY" LIKE
'%FR%' AND ("TYPE"='A' OR "TYPE"='C')) |
Si quelqu'un a une piste....