High-Availability Configurations
High-availability configurations are similar to other configurations of Microsoft SQL Server Notification Services, except that they use clustering.
Failover Clustering
A failover cluster is a group of independent servers that run Microsoft Cluster Service and work collectively as a single system. The purpose of a failover cluster is to keep applications running during server failures and planned outages. If one of the servers in the cluster becomes unavailable, resources and applications move to another available cluster node.
For an effective Notification Services clustering solution, you must cluster the databases and Notification Services.
The first failover clustering illustration shows the clustered version of a single-server configuration. Notification Services and the databases are on the same node. If one node fails, the other node picks up all processing. External applications, such as subscription management applications, connect to a virtual server name that represents the cluster.
.gif "Clustered single-server configuration")
The second failover clustering illustration shows the clustered version of the remote database configuration. In this example, the databases are on a two-node failover cluster, and Notification Services also uses a two-node failover cluster. External applications, such as subscription management applications, connect to SQL Server using the virtual name for the SQL Server failover cluster.
.gif "Clustered remote-server configuration")
Other High-Availability Configurations
You also can use other high-availability options provided by SQL Server, such as log shipping. For more information, see Maintaining High Availability.
See Also
Tasks
Walkthrough: Clustered Single-Server Deployment
Concepts
Hardware Configurations
Notification Services Considerations
Database Considerations
Database Resource Planning