Re: Make the planner smarter about idiosyncratic (or broken) ORM behaviour

From: Simon Riggs <simon(at)2ndQuadrant(dot)com>
To: Frank Jördens <frank(at)woome(dot)com>
Cc: pgsql-sql(at)postgresql(dot)org, Nic Ferrier <nferrier(at)woome(dot)com>, Edward George <ed(at)woome(dot)com>, Seb Potter <seb(at)woome(dot)com>, Hannu Krosing <hannu(at)2ndquadrant(dot)com>
Subject: Re: Make the planner smarter about idiosyncratic (or broken) ORM behaviour
Date: 2009-05-22 10:40:03
Message-ID: 1242988803.27960.382.camel@ebony.2ndQuadrant
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-sql


On Fri, 2009-05-22 at 10:33 +0100, Frank Jördens wrote:
> Almost all of the 8k queries per second that are hitting our database
> are generated by a web development framework's ORM (Django), and some
> of the syntax there is quite mad. Possibly the worst tic exhibited by
> the translation layer is that if you filter a query on an object by
> more than one property on a related object it will create a join for
> each property:

It's a known issue that we're looking to work on in the next release.

Work started in 8.4, though was more complex than first appeared and we
didn't complete it in time for the dev deadline.

--
Simon Riggs www.2ndQuadrant.com
PostgreSQL Training, Services and Support

In response to

Browse pgsql-sql by date

  From Date Subject
Next Message Isaac Dover 2009-05-22 13:03:01 Re: help with pg_hba.conf
Previous Message Richard Huxton 2009-05-22 10:33:35 Re: Make the planner smarter about idiosyncratic (or broken) ORM behaviour