Re: For index bloat: VACUUM ANALYZE vs REINDEX/CLUSTER

Поиск
Список
Период
Сортировка
От Richard Broersma Jr
Тема Re: For index bloat: VACUUM ANALYZE vs REINDEX/CLUSTER
Дата
Msg-id 162380.54690.qm@web31806.mail.mud.yahoo.com
обсуждение исходный текст
Ответ на Re: For index bloat: VACUUM ANALYZE vs REINDEX/CLUSTER  (Erik Jones <erik@myemma.com>)
Ответы Re: For index bloat: VACUUM ANALYZE vs REINDEX/CLUSTER  (Erik Jones <erik@myemma.com>)
Список pgsql-general
> Not quite.  Once a page has reached it's fill factor percentage full,
> no more inserts will happen on that page, only updates.  Also, I
> think you have large/small backwards wrt fill factor.  If  you have a
> fill factor of, say, 40% then once a page has reached 40% full no
> more inserts will happen (unless some space is reclaimed by vacuum).
> So, smaller fill factors == bigger holes.  The bigger the fill
> factor, the smaller the whole:  if you have a fill factor of 90%,
> only 10% is reserved for updates of rows on that page.

So (just to reiterate), fill factor can be applied to both a table and/or an index(es).  But the
"holes" built into the page of a table or index can only be filled by UPDATE Statements.

Thanks for the clarification!

Regards,

Richard Broesma Jr.


В списке pgsql-general по дате отправления:

Предыдущее
От: Erik Jones
Дата:
Сообщение: Re: For index bloat: VACUUM ANALYZE vs REINDEX/CLUSTER
Следующее
От: Decibel!
Дата:
Сообщение: Re: Is this good spec for a PostgreSQL server?