Re: transparent use of pre-aggregated value for very large datasets

From: Greg Stark <gsstark(at)mit(dot)edu>
To: pgsql-general(at)postgresql(dot)org
Subject: Re: transparent use of pre-aggregated value for very large datasets
Date: 2003-03-03 20:36:32
Message-ID: 87bs0sjigf.fsf@stark.dyndns.tv
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general


chaudhar(at)umich(dot)edu (shahbaz) writes:

> Any way, while I could tune the heck out of the DB or spend countless
> thousands on RAM, wouldn't it be better to take advantage of cheap
> hard-disk space.

So the main objection people have had to such caching is that it would
conflict with the concurrency goals. The aggregate caches would become a big
point of contention. There are a lot of applications where there are a lot
more selects than update/inserts however. And there are a lot of applications
where some information doesn't need to be updated continuously.

It's kind of sad to see the rules system not quite being up to what you need.
Are you sure it can't be done? It would be the cleanest way to do it manually.

--
greg

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Cristian Custodio 2003-03-03 20:53:45 Array like a foreign key
Previous Message Dennis Gearon 2003-03-03 20:31:36 Re: almost there on a design