Multiple Tables and dupes

Hello,

I am soon going to break up our 18-21 month history database into 7 3-month pieces to make the 'purge' process much quicker by deleting the oldest 3 months of data.

Our database has tables that contain 18 months of 'base' data and 3 months of 'incremental' data (all the same layout).

Currently, when we insert into the incremental table (the base is read-only), a trigger is activated to make sure the row doesn't already exists on the base (can happen at times) and the row isn't inserted.

By breaking the table(s) into 7 pieces, what's the best method of preventing duplicate rows from being added?

We're on Oracle 7.1.4, to upgrade to 7.3.2.x soon, so I know I can use Partition Views for the queries, but am now a bit concerned on the performance side of inserts, as it has to check against the other 6 pieces of the table.

Any comments, ideas?

Thank you!

-Bruce Received on Sun Apr 20 1997 - 00:00:00 CDT