From: | Guillaume Lelarge <guillaume(at)lelarge(dot)info> |
---|---|
To: | PegoraroF10 <marcos(at)f10(dot)com(dot)br> |
Cc: | PostgreSQL General <pgsql-general(at)postgresql(dot)org> |
Subject: | Re: Cluster and Vacuum Full |
Date: | 2020-10-05 11:14:39 |
Message-ID: | CAECtzeWAx+GOHaj58YoGFphVK22ahykBRVuPqi17efODGJM+5A@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Le lun. 5 oct. 2020 à 12:22, PegoraroF10 <marcos(at)f10(dot)com(dot)br> a écrit :
> I have tables which are Master Detail and usually my program loads all
> detail
> records of a master record. So I configured CLUSTER on all those detail
> tables to use an index which is the relation with master table. With that I
> can load less records to Shared Buffers because all detaild records are on
> sequencial pages, that´s fine.
>
> I know that to have this working I have to periodically run CLUSTER, ok
> too.
>
> But today instead of CLUSTER I just ran VACUUM FULL. How vacuum full
> recreates entirelly that table, like cluster does, I thought it would use
> that index too, but didn´t. I didn´t see what sequence of records it did
> but
> was not the one I´ve configured with cluster.
>
> Is that expected or Vacuum should obbey cluster configuration ?
>
>
Only CLUSTER obeys cluster configuration. VACUUM FULL in particular doesn't.
--
Guillaume.
From | Date | Subject | |
---|---|---|---|
Next Message | Mario Emmenlauer | 2020-10-05 11:22:18 | dup(0) fails on Ubuntu 20.04 and macOS 10.15 with 13.0 |
Previous Message | Han Parker | 2020-10-05 10:08:28 | 回复: May "PostgreSQL server side GB18030 character set support" reconsidered? |