Re: <= Index.
От | Greg Sikorski |
---|---|
Тема | Re: <= Index. |
Дата | |
Msg-id | OF12E0FEFE.59B34BA8-ON80256B8D.0050F0AD-80256B8D.00515EF1@atomicrevs.net обсуждение исходный текст |
Ответ на | <= Index. ("Greg Sikorski" <gte@atomicrevs.demon.co.uk>) |
Ответы |
Re: <= Index.
|
Список | pgsql-admin |
pgsql-admin-owner@postgresql.org wrote on 31/03/2002 08:06:52: > Greg Sikorski writes: > > > How come this Index isn't being used in the situation below? > > There's no point in using an index if you're retrieving nearly the entire > table. > > -- > Peter Eisentraut peter_e@gmx.net > > Well, that was probably a poor example to include, but it was nearly 4am ;) Typically <20 results are returned, although it still says "Seq Scan", but on a smaller subset. -- cmaster=# SELECT user_id,channel_id FROM levels WHERE suspend_expires <= 1017589362 AND suspend_expires <> 0; user_id | channel_id ---------+------------ ... (17 rows) cmaster=# explain analyze SELECT user_id,channel_id FROM levels WHERE suspend_expires <= 1017589362 AND suspend_expires <> 0; NOTICE: QUERY PLAN: Seq Scan on levels (cost=0.00..13709.09 rows=2609 width=8) (actual time=208.98..1521.08 rows=17 loops=1) Total runtime: 1521.29 msec Cheers, Greg.
В списке pgsql-admin по дате отправления: