Re: experience with tags schemas - Mailing list pgsql-general

From Chris Browne
Subject Re: experience with tags schemas
Date
Msg-id 87y697r0jf.fsf@cbbrowne.afilias-int.info
Whole thread Raw
In response to experience with tags schemas  (Matthieu Huin <matthieu.huin@wallix.com>)
List pgsql-general
matthieu.huin@wallix.com (Matthieu Huin) writes:
> Greetings,
>
> Does anyone here have any experience with tags schemas on postgresql ?
>
> I am struggling with a schema inspired by "scuttle" described here :
>
> http://www.pui.ch/phred/archives/2005/04/tags-database-schemas.html
>
> Performances drop dramatically when the data set increases. Is it
> "normal" and I should look at other data structures ? Or am I doing
> something wrong ?
>
> If you need more information, I'll be happy to give it to you.

I'd expect the "Toxi" solution (where do they get the names!??!?!) to be
the fastest one of the litter, as most of the "burden" of work goes into
the narrow table, "tagmap."  (It's also easily extended to cope with the
way Deli.cio.us does tagging, as you can attach user IDs to "tagmap")

In any case, it would make a lot of sense to run EXPLAIN ANALYZE on some
of the queries; that ought to provide some insight as to what parts are
behaving notably badly.

The article provides some good grist for the argument that proper
normalization frequently *IMPROVES* performance.  (Contrary to the all
too common "NoSQL is Web Scale" arguments that we must "denormalize for
performance.")
--
If this was helpful, <http://svcs.affero.net/rm.php?r=cbbrowne> rate me
http://www3.sympatico.ca/cbbrowne/lsf.html
"Like I've always  said, if you don't have anything  nice to say, come
sit by me." -- Steel Magnolias

pgsql-general by date:

Previous
From: Dann Corbit
Date:
Subject: Re: Access postgresql data base from .net
Next
From: Dimitri Fontaine
Date:
Subject: Re: Cloning database without dump/restore