Re: max time in a table query takes ages
| От | Grzegorz Jaśkiewicz |
|---|---|
| Тема | Re: max time in a table query takes ages |
| Дата | |
| Msg-id | 2f4958ff0810230652m5fcce9dcva0b98cf854eff76b@mail.gmail.com обсуждение исходный текст |
| Ответ на | Re: max time in a table query takes ages (Tom Lane <tgl@sss.pgh.pa.us>) |
| Ответы |
Re: max time in a table query takes ages
|
| Список | pgsql-general |
On Thu, Oct 23, 2008 at 2:47 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
--
GJ
I suppose the problem is that rows with processed = false are very few
in the upper range of data. If so, and if you really need this to go
fast, a partial index might be worth its overhead:
create index foo on staticstats(data) where processed = false;
I mostly scan that table for processed=false. Everything with processed = true is really useless, and there to guard other tables against duplication (all other tables contain pretty much the same data, but information about relation isn't stored - because whole structure is quite complicated, and I need to retrieve data from other tables in matter of miliseconds - I get live graphs plotted based on huge range of input params, no chance to pre-render it).
I also thought about create index bar staticstats(data, processed) where processed = false;
but that would be useful only to that query.
Can postgres use combined indicies for queries that would only require part of it ?
iow, do I have to create index for every possible combination that can be used in a query ? what's the best way to do it ?
--
GJ
В списке pgsql-general по дате отправления: