Skip site navigation (1)
Skip section navigation (2)
Peripheral Links
Header And Logo
|
Site Navigation
Re: Feature suggestion : FAST CLUSTER
- From: PFC <lists(at)peufeu(dot)com>
- To: "Alexander Staubo" <alex(at)purefiction(dot)net>
- Cc: "Jim C. Nasby" <decibel(at)decibel(dot)org>, "pgsql-performance(at)postgresql(dot)org" <pgsql-performance(at)postgresql(dot)org>
- Subject: Re: Feature suggestion : FAST CLUSTER
- Date: Sun, 27 May 2007 20:44:20 +0200
How does it know what to cluster by? Does it gather statistics about query patterns on which it can decide an optimal clustering, or does it merely follow a clustering previously set up by the user?
Nothing fancy, InnoDB ALWAYS clusters on the primary key, whatever it is. So, if you can hack your stuff into having a primary key that clusters nicely, good for you. If not, well... So, I used (topic_id, post_id) as the PK, even though it isn't the real PK (this should be post_id)...
- References:
- Feature suggestion : FAST CLUSTER
- From: PFC
- Re: Feature suggestion : FAST CLUSTER
- From: Jim C. Nasby
- Re: Feature suggestion : FAST CLUSTER
- From: PFC
- Re: Feature suggestion : FAST CLUSTER
- From: Alexander Staubo
- Feature suggestion : FAST CLUSTER
- Prev by Date: Re: Feature suggestion : FAST CLUSTER
- Next by Date: Re: Postgres Benchmark Results
- Previous by thread: Re: Feature suggestion : FAST CLUSTER
- Next by thread: Re: Feature suggestion : FAST CLUSTER
- Indexes: