Re: Indexes on inheriting tables - Mailing list pgsql-general

From Scott Mead
Subject Re: Indexes on inheriting tables
Date
Msg-id CAKq0gvJ89_OWr1GJytrtX=X9YNRXxWoPJ-kDq3EBSevaMaga4g@mail.gmail.com
Whole thread Raw
In response to Re: Indexes on inheriting tables  (Ondrej Ivanič <ondrej.ivanic@gmail.com>)
List pgsql-general

2011/8/24 Ondrej Ivanič <ondrej.ivanic@gmail.com>
Hi,

On 25 August 2011 11:17, Toby Corkindale
<toby.corkindale@strategicdata.com.au> wrote:
> Do I need to make sure I re-create every index on every child table I
> create?
> That would be.. annoying, at best.

Yes, it is little bit annoying but I like it. You don't need any index
on parent table but you have to create them "manually". I wrote simple
python script which creates partitions and required indexes in advance
(tables are partitioned by date).

I like the flexibility because you can have different indexex on
different partitions. For example, I discovered that adding index will
improve several queries. In the production I can't afford exclusive
lock (build index concurrently takes ages) so I updated and re-run the
script which re-created future partitions.


My Personal favorite is the LIKE syntax:

  CREATE TABLE foo_1 (LIKE foo including indexes ) inherits (foo);

  It doesn't help you change children after the fact, but your new partitions automatically get whatever indexes you've stuck on the master table.

--Scott
 

--
Ondrej Ivanic
(ondrej.ivanic@gmail.com)

--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general

pgsql-general by date:

Previous
From: Alan Hodgson
Date:
Subject: Re: heavy swapping, not sure why
Next
From: Tom Lane
Date:
Subject: Re: heavy swapping, not sure why