On Wed, 28 Oct 2009 10:12:19 -0500 Peter Hunsberger <peter.hunsber...@gmail.com> wrote:
> > The first approach requires a distinct/group by that may be > > expensive. > > The second one requires I keep in memory all the emails while the > > first statement run. > Unless you're dealing with 100,000's of these things I think you're > engaging in a process of "premature optimization". Group by can > work efficiently over millions of rows. We may get in the range of half that number occasionally but not feeding emails directly from a HTTP request. Still the number of passwords generated in one run may be in the range of 50K. But well I could calmly wait 2 or 3 seconds. Making some very rough test on a similar box to the one I'll have to use it takes few milliseconds on a not indexed table. > Do the simplest thing possible. Get it working, then see if you > have any new problems you need to solve. Every issue you've > described so far is database design 101 and should present no real > problem. I think you're agonizing over nothing... That's always a good advice. Sometimes you're out just for moral support. thanks -- Ivan Sergio Borgonovo http://www.webthatworks.it -- Sent via pgsql-general mailing list (pgsql-general@postgresql.org) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-general
