DB-A will be chosen as the Primary Node responsible for real time and multi-theaded data synchronization to DB-B. There are always two real-time consistent and alive datasets which enhances the reliability of the whole system. In case DB-A is down, DB-B will remain running or vice versa with Zero Data Loss.

      DBTwin Active-Active Database Cluster Diagram

For more details about the DBTwin synchronization
https://www.zhetides.com/nd.jsp?id=36

Why Always On doesn't have Guaranteed Zero Data Loss(GZDL)?

https://www.zhetides.com/nd.jsp?id=24