Re: Inefficient query plan - Mailing list pgsql-performance

From Grzegorz Jaśkiewicz
Subject Re: Inefficient query plan
Date
Msg-id AANLkTi=iBxk7s0jD5biB58LQRGWMNJy6E7cQrw3DyHwO@mail.gmail.com
Whole thread Raw
In response to Re: Inefficient query plan  ("Kevin Grittner" <Kevin.Grittner@wicourts.gov>)
Responses Re: Inefficient query plan  ("Kevin Grittner" <Kevin.Grittner@wicourts.gov>)
List pgsql-performance
On Mon, Aug 23, 2010 at 2:47 PM, Kevin Grittner
<Kevin.Grittner@wicourts.gov> wrote:
> Grzegorz Jaœkiewicz<gryzman@gmail.com> wrote:
>
>> joining on varchars is always going to be very expensive. Longer
>> the value is, more expensive it will be. Consider going for
>> surrogate keys.
>
> Surrogate keys come with their own set of costs and introduce quite
> a few problems of their own.  I don't want to start a flame war or
> go into an overly long diatribe on the evils of surrogate keys on
> this thread; suffice it to say that it's not the first thing to try
> here.
>
> As an example of the performance we get using natural keys, with
> compound keys on almost every table, check out this 1.3TB database,
> being updated live by 3000 users as you view it:
>
> http://wcca.wicourts.gov/
>
> Some tables have hundreds of millions of rows.  No partitioning.
>

True, but as far as joining is concerned, joining on single column
fixed length fields is always going to be a win. Hence why surrogate
keys make sens in this particular example, or the guy here should at
least test it to see, rather than believe in one or the other.


--
GJ

pgsql-performance by date:

Previous
From: "Kevin Grittner"
Date:
Subject: Re: Inefficient query plan
Next
From: "Kevin Grittner"
Date:
Subject: Re: Inefficient query plan