Basic SVC PPRC cluster configuration and operation

In a typical setup with AIX® server nodes in a wide-area PowerHA® SystemMirror® cluster configuration, the server nodes at each PowerHA SystemMirror cluster site are connected directly to the local SVC Clusters. Two or more inter-cluster links are required between the two SVC clusters. The PowerHA SystemMirror nodes at the same geographical site access the same shared volume groups, but the nodes at each site access them from different physical volumes.

The SVC PPRC maintains separate identical local copies of the application data on two separate back end storage subsystems. Several virtual disks are mirrored via primary SVC cluster PPRC from the primary site to the standby site. When node or site failure occurs, access to disk resources are not passed from one node to another. Instead, all highly available applications are restarted at the standby site using data copy on secondary volumes. Under normal operation, the application is active on a server at the production site, and all updates to the application data are automatically replicated to the backup disk subsystem by the SVC PPRC framework. PPRC protects the backup copy from inadvertent modification. When a total site failure occurs, the application is restarted on a backup server at the remote site.

Prior to restarting the application, which, in this context, means not only the application that end users interact with, but also all dependent data base software or other middleware, PowerHA SystemMirror Enterprise Edition initiates SVC actions to ensure that the backup disk volumes are in the appropriate state to allow application access.

PowerHA SystemMirror support for geographic clustering is based on the concept of replicated resources . These are defined as a resource type that has a primary and secondary instance, corresponding to the source and target of data copies that are replicated across two locations. In SVC PPRC, a PPRC consistency group, comprised of a list of SVC relationships defined to SVC and associated with an PowerHA SystemMirror geographically separated cluster definition and defined to PowerHA SystemMirror is referred to as an SVC PPRC replicated resource . The definition for an SVC PPRC replicated resource contains the VDisk name and the PowerHA SystemMirror volume groups associated with the VDisk pairs. SVC recognizes which volumes mirror each other for each PPRC replicated resource.

In the example shown here, the PowerHA SystemMirror production site includes:

• Server A and Server B
• The ESS labeled Primary ESS
• A SAN Volume Controller connected by Fibre Channel PPRC links to the SVC at the secondary site and two connections to the Primary ESS.

The PowerHA SystemMirror recovery site includes:

• Server C and Server D
• The ESS labeled Secondary ESS
• A SAN Volume Controller connected by Fibre Channel PPRC links to the SVC at the primary site and two connections to the Primary ESS.

The PowerHA SystemMirror resource group contains:

• The four server nodes
• IP networks to connect the server nodes
• One or more shared volume groups
• The PPRC replicated resource associated with the volumes in the volume group.

The configuration also includes point-to-point networks for heartbeating to connect the cluster nodes.

Note that all nodes also require connections to the SVC. At least one storage controller is connected to each SVC cluster via a SCSI or Fiber Channel connection.