Hi Mark,

Thanks for writing this article it was interesting to see a different way of implementing CDC incremental loading. I have also implemented an incremental load using CDC and I found that the really important part is ensuring synchronisation between the data warehouse tables and the tracked tables. You mention a couple of methods for resynchronisation in your article and I have an additional method which works quite reliably for me which I thought you might be interested in.

In my solution I have a SQL replication step between the production OLTP and the data warehouse. This gives me control over the data flow into the monitored tables and allows me to pause replication apply changes to the data warehouse. Once finished I can then restart the data flow and CDC monitoring without losing any of the change data. This of course requires the replication subscriptions to be setup not to expire during the stoppage.

Cheers

Martin