Re: sql question (hopefully) - Mailing list pgsql-novice

Josh,

Thanks.  I created a plpgsql function that works but I'll take your advice
and do it in C.

Hopefully, I'll find a way to justify the hiring of true DBA.  It has always
been a *challenge* for the companies I've worked for because DBA = cost.  Oh
well, maybe it's because I got the job done anyhow from their point of view.
=)  Hmmm.. welcome to the Third World.

I was actually made to comply to adopt the darned schema because of the
faster reply despite the setbacks I'm (we were) well aware of.  We're in
this business where we could get a million hits in a span of a few minutes
and we need to reply to these hits ASAP or we lose more potential traffic
which meant lost potential revenue.  Yes, we earn from the traffic.

The SQL solution would be good stock knowledge for future reference but then
again, hopefully I wouldn't have to use it anymore. =)

With the help of the Postgres lists, the documentation that's getting better
and better, ArsDigita's and now RedHat's OpenACS (and in the future the
Postgres books I've already ordered), I've made tremendous progress from 0
knowledge in RDBMS specifically in Postgres.  So, thanks to you guys and
hopefully you don't get tired of replying to the questions because they sure
help a lot.

--
Mel

p.s.  I'm hoping to publish the schema here in the near future to get your
opinion on how to improve it without sacrificing too much speed.


-----Original Message-----
From: pgsql-novice-owner@postgresql.org
[mailto:pgsql-novice-owner@postgresql.org]On Behalf Of Josh Berkus
Sent: Wednesday, June 18, 2003 11:14 PM
To: mel@gmanmi.tv; 'Ron Johnson'; 'PgSQL Novice ML'; 'Dani Oderbolz'
Subject: Re: [NOVICE] sql question (hopefully)


Mel,

> i wanted to find out for sure if i really won't get the desired result(s)
> in sql and have to create a function instead.

Perhaps, but the function better be in C, or your performance will continue
to
suck ...

> yes it breaks the 1st normal form.  we did this for some *practical*
> purposes supposedly. =) i would've wanted to discuss it here but i'll
> probably bore you to death or i'd end up starting a new discussion
> altogether.  since it's in production and i need to do a quick patch
> without touching the external programs depending on this schema, i've to
> find a quick remedy.

To be frank, boyo, you're on the road to hell.

I am constantly amazed at how many newbie DBAs adopt not-normalized schema
because "it's faster" and then have to add in workarounds that kill the
performance of their database.

Anyway, if someone else doesn't solve it, I'll come up with a SQL solution
later in the week ....

--
Josh Berkus
Aglio Database Solutions
San Francisco

---------------------------(end of broadcast)---------------------------
TIP 6: Have you searched our list archives?

               http://archives.postgresql.org