DB2 Replication: Change Aggregate Table

Hi all,

I have a small problem with using DB2 V8.1.4 Replication Center.
Actually the problem is that I have not been able to figure out what
exactly is a change aggregate table and its use. It would be really
helpful if anybody could help me out by providing a simple example on
how the change aggregate table may be used to repliate changes that
occur at the source table.

Thanks in advance,
Varun Jain

I've never used the replication feature in DB2. But from what I've read, a Change Aggregate table is used with a Base Aggregate table to allow statistcal analysis of data. An example might be that you would run some computations on your source data (percentages, sums, averages, etc.) . You would store this snapshot in the Base Aggregate table. As changes are made to your data, you replicate the changes to your Change Aggregate table. Then if you wanted to see totals and other values in "real time" it would just be a matter of calculating the values of the baseline snapshot in Base Aggregate table plus the recently added values in the Change Aggregate table.

A setup like this would prevent you from having to recalculate millions/billions of rows of data each time a value was added to the database. At regular intervals (nightly/weekly/monthly), you would empty your Change Aggregate table and recalculate your Base Aggregate table and start over again.