From: | Alex Dovlecel <dovle(at)kbs(dot)twi(dot)tudelft(dot)nl> |
---|---|
To: | pgsql-jdbc(at)postgresql(dot)org |
Subject: | Multiple queries OR one and big |
Date: | 2002-12-02 16:46:27 |
Message-ID: | E18ItiF-0002aB-00@ford.kbs.twi.tudelft.nl |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-jdbc |
Hello all,
This is a little bit out of topic but is related to jdbc.
I have a database containing three tables : Route, RouteItem and
RouteItemStop (Route : 1 -- n RouteItems 1 -- n RouteItemStops ). Hope you
see the picture. :o)
I want to obtain a route with a specific ID. I have two ways of implementing
it:
1) one big sql join statement that links all those 3 tables. The problem will
be the huge ammount of data (like all the Route data will be repeated for n^2
times , the same data). The advantage: only one query.
2) join by route and routeitem and then get all the routeItemStops for each
route item. This means n+1 queries. Advantage : don't have all that ammount
of inutile data sent from sql server. Disadvantage: n+1 sql statements to be
executed, instead of only one.
Is there any 'better' solution. If not, what should I choose? (don't have the
experience and I don't know how psql sends a join query: optimised, not
optimised)...
For now join is just Where route.id = route_item.route_id.
If I am going to use the JOIN sql command, will that be faster and better?
Why ?
Tx
dovle
From | Date | Subject | |
---|---|---|---|
Next Message | David Hooker | 2002-12-02 16:51:57 | Updatable result sets |
Previous Message | Rasputin | 2002-12-02 16:21:34 | Re: md5 passwords in 7.2.3 ? |