dm.aeqa@gmail.com (DM) writes:
> It is not real time, updates every 5 mins should be fine.
>
> But the DB2 database is real busy and its real performance based.
The book "Scalable Internet Architectures" (by Theo Schlossnagle) has
an example of how to build a trigger-based replication system copying
data from an Oracle database to Postgres.
It basically tracks PK values for tuples changed/deleted (which is
what the old RServer and eRServer replication systems for Postgres
did), allowing a process to come in afterwards and pull data over to
the replica.
I presume that DB2 has enough functionality to let you run triggers to
capture which tuples changed, and when. Given that, it shouldn't be
super-difficult to do what you need.
--
select 'cbbrowne' || '@' || 'cbbrowne.com';
http://cbbrowne.com/info/slony.html
"MS apparently now has a team dedicated to tracking problems with
Linux and publicizing them. I guess eventually they'll figure out
this back fires... ;)" -- William Burrow <aa126@DELETE.fan.nb.ca>