Introduction and considerations

The Microsoft Cluster Service (MSCS) allows you to configure two-node high availability clusters on standard PC server hardware. MSCS server clusters provide redundancy and fault resilience for the WebSphere and application data services. The Microsoft Windows NT® Server Enterprise Edition and the Windows 2000 Advanced Server include MSCS.

MSCS uses a share-nothing clustering architecture, which has no concurrently shared resources. It works by transferring the ownership of resources from one node to another automatically to keep data services available in case of failures. You can also manually switch from one node to another using operator commands. MSCS provides the ability to group resources and express dependencies between them, control process starting and stopping in a logical sequence, and move them from one node to another. It is also possible to configure a failback policy that controls whether an application moves back to its former node when that node becomes available again.

WebSphere is configured to use virtual IP addresses that are under MSCS cluster control, as shown in Figure 12-9. When database services (DB2, Oracle, MS SQL Server) move from one clustered node to the other, it takes its virtual IP address with it. The virtual IP address is different from the stationary physical IP address. The virtual IP address make database failover transparent to WebSphere and other application clients.

Figure 12-9 Microsoft Cluster Service configuration for WebSphere

  Prev | Home | Next

 

WebSphere is a trademark of the IBM Corporation in the United States, other countries, or both.

 

IBM is a trademark of the IBM Corporation in the United States, other countries, or both.