August 19, 2014 at 2:58 pm
Is there a way for me to set up CDC so that all the processing (SQL Agent, etc) happens on the machine receiving the data? I'd like to move as much of the processing as possible to the destination.
TIA
Andre
October 16, 2014 at 6:07 pm
Hi, I am not clear on your context here. CDC runs on a database where the data is being monitored for changes. How does this relate to your "destination "?
----------------------------------------------------
October 17, 2014 at 9:42 am
Not so much. It seems that all the Change Data Capture stuff is a "push" to the destination.
I was hoping that it would be more like replication where the Subscriber can "pull" from the Publisher. I wanted to take as much of the processing work as possible away from the source server.
October 17, 2014 at 12:33 pm
You'll have to tell me about how this is a push to the destination. Normally you query the change data capture tables to see what rows changed and get the latest change (if it changed more than once). From here you can push/pull this elsewhere. Is this what you mean?
A ETL to another system is normally ran on a periodic basis, so you are correct here. If you need the data in one system to automatically update to another in real time then you are talking transactional replication.
What type of environment are you pushing to? What is the overall goal as I am getting the impression CDC is not what you want? CDC's intention is to aid in warehouse maintenance by not pulling records into a warehouse that are already there from the production system and unchanged from the last ETL run.
----------------------------------------------------
Viewing 4 posts - 1 through 3 (of 3 total)
You must be logged in to reply to this topic. Login to reply