Re: [GENERAL] Large data and slow queries
От | vinny |
---|---|
Тема | Re: [GENERAL] Large data and slow queries |
Дата | |
Msg-id | 45aacae7eeaa24e4d5559028458c3f3d@xs4all.nl обсуждение исходный текст |
Ответ на | Re: [GENERAL] Large data and slow queries ("Martijn Tonies \(Upscene Productions\)" <m.tonies@upscene.com>) |
Ответы |
Re: [GENERAL] Large data and slow queries
|
Список | pgsql-general |
On 2017-04-19 13:25, Martijn Tonies (Upscene Productions) wrote: > Samuel, others, > > Perhaps I'm missing something, but I'd be interested in the reasoning > behind this. > > For column 'what', it seems you have no index on all values, only > indices with specific values for 'what'. > > How does this speed up the search? Will PostgreSQL use those indices, > instead of using a generic index on 'what' and optionally other > columns? > > > With regards, > > Martijn Tonies > Upscene Productions > http://www.upscene.com > That's a "partial index", it only contains records that meet the requirements of the index definition. https://www.postgresql.org/docs/9.5/static/indexes-partial.html Basically; if you create an index on records where 'name = kees' then if your query contains "where name=kees" the planner can just load that index and know that the records in that index will not contain any other names, saving the need to filter for 'name=kees'
В списке pgsql-general по дате отправления: