Re: PG using index+filter instead only use index

От: Tom Lane
Тема: Re: PG using index+filter instead only use index
Дата: ,
Msg-id: 5135.1269028162@sss.pgh.pa.us
(см: обсуждение, исходный текст)
Ответ на: PG using index+filter instead only use index  (Alexandre de Arruda Paes)
Ответы: Re: PG using index+filter instead only use index  (Alexandre de Arruda Paes)
Список: pgsql-performance

Скрыть дерево обсуждения

PG using index+filter instead only use index  (Alexandre de Arruda Paes, )
 Re: PG using index+filter instead only use index  (Tom Lane, )
  Re: PG using index+filter instead only use index  (Alexandre de Arruda Paes, )
   Re: PG using index+filter instead only use index  (Tom Lane, )

Alexandre de Arruda Paes <> writes:
> My question: if the cost is exactly the same, why PG choose the index
> ict13t2 on ct13t and apply a filter instead use the primary key ?

Why shouldn't it, if the estimated costs are the same?  You didn't
actually demonstrate they're the same though.

The cost estimates look a bit unusual to me; are you using nondefault
cost parameters, and if so what are they?

            regards, tom lane


В списке pgsql-performance по дате сообщения:

От: Kenneth Marshall
Дата:
Сообщение: Re: GiST index performance
От: Tom Lane
Дата:
Сообщение: Re: PG using index+filter instead only use index