Re: Plan time Improvement - 64bit bitmapset
От | Andres Freund |
---|---|
Тема | Re: Plan time Improvement - 64bit bitmapset |
Дата | |
Msg-id | 4A26F7A5.6060800@anarazel.de обсуждение исходный текст |
Ответ на | Re: Plan time Improvement - 64bit bitmapset (Gregory Stark <stark@enterprisedb.com>) |
Список | pgsql-hackers |
On 06/03/2009 08:57 PM, Gregory Stark wrote: > "Kevin Grittner"<Kevin.Grittner@wicourts.gov> writes: >> Andres Freund<andres@anarazel.de> wrote: >>> long plan times (minutes and up) >> Wow. I thought I had some pretty complex queries, including some >> which join using several views, each of which has several joins; >> but I've never gone to multiple seconds on plan time (much less >> multiple minutes!) without very high statistics targets and many >> indexes on the tables. Any rough estimates on those? > My money's still on very large statistics targets. If you have a lot > of columns and 1,000-element arrays for each column that can get big > pretty quickly. Only a relatively small difference (stat target 10; 1000): 22283.187 23986.504 > But that doesn't explain the bitmap ops being important. Hm. Actually > having a lot of columns and then joining a lot of tables could mean > having fairly large bitmapsets. Some of the views have a large amount of columns (200-400) - none of the actual tables has many columns though (10 user columns is the biggest I think). 110 tables containing real data. The queries include the same tables quite often. Andres
В списке pgsql-hackers по дате отправления: