AlwaysOn Availability Groups are configured with the availability group wizard. In the sample design shown in the following figure, the availability group is configured with a primary replica database (MS-SQL-DB1\PROD01) and one local secondary read-only replica (MS-SQL-DB2\PROD01) at the customer’s on-premises data center. Two secondary replicas (MS-SQL-DB3\PROD01 and MS-SQL-DB4\PROD01) are configured at the VMware Cloud Provider facility for high availability and disaster recovery purposes. This architecture provides redundancy in multiple failure scenarios at both the on-premises and provider data centers.

The automatic failover mode option provides high availability by causing the database to fail over quickly and making it available after the loss of the primary replica. In the event of an unplanned outage, the failover takes place without manual intervention by operational teams or database administrators.

To configure an AlwaysOn Availability Group for automatic failover, the current primary replica of the database and one secondary replica must be set to synchronous-commit mode with automatic failover. Synchronized mode provides for little or no data loss, but requires more available network bandwidth, and therefore the environment must be within the minimum latency requirement limits set by Microsoft. In a typical architecture, to maximize performance and minimize data loss for disaster recovery purposes, configure all replicas for synchronous-commit mode with automatic failover.