Re: Performance issues of one vs. two split tables.
От | Bill Moseley |
---|---|
Тема | Re: Performance issues of one vs. two split tables. |
Дата | |
Msg-id | 20070516060106.GE19117@hank.org обсуждение исходный текст |
Ответ на | Re: Performance issues of one vs. two split tables. (PFC <lists@peufeu.com>) |
Список | pgsql-general |
On Tue, May 15, 2007 at 09:20:53PM +0200, PFC wrote: > >From an outside perspective it just > >seems odd that potentially a large amount of data would be pulled off > >disk into memory that is never used. Perhaps there's an overriding > >reason for this. > > Yeah, where would you put this data if you didn't put it where it is > now ? Swish-e isn't a database by any means, but it does have a way to store column like meta data for each "row". When it does a search it only explicitly pulls from disk the meta data that it's asked to return. Granted, the OS is reading from disk more than the application is asking for, but the application is only allocating memory for the data it's going to return. And the column (meta data) is not always stored together on disk. Without knowing Pg internals I wasn't aware of how the actual table data was organized and fetched into memory. > "Premature optimization is the root of all evil" Exactly what prompted this thread. ;) -- Bill Moseley moseley@hank.org
В списке pgsql-general по дате отправления: