The logical partition (LPAR) which is a Cluster Server (VCS) node and controls other LPARs on the physical server is referred to as a management LPAR (MPLAR).
VCS monitors and manages LPARs using LPAR resources.
VCS enables the following for management LPARs:
Connects multiple LPARs to form a cluster for increased availability
Redundant Hardware Management Console (HMC) support
Multiple VIOS support
Enables nodes to cooperate at the software level to form a cluster
Enables other nodes to take predefined actions when a monitored application fails, for instance to take over and bring up applications elsewhere in the cluster
Enables monitoring and failover of LPARs configured in VCS as LPAR resources
The VCS LPAR agent supports redundant HMC configurations. VCS can use any HMC which is up and running to manage and monitor the LPARs.
Multiple VIOS support provides high availability to LPARs in case of VIO server(s) crash: LPAR agent provides high availability against VIO server(s) crash for the managed LPARs. If all the VIO servers specified are down, managed LPARs are failed over to another host.
For more information on Cluster Server features, see the Cluster Server Administrator's Guide.