Anthony Hull |
CLIQUES
The clique is a feature of multinode systems that physically groups nodes together by muliported access to common disk array units. Inter-Node disk array connections are made using SCSI buses.
Shared SCSI-II paths enable redundancy to ensure that loss of a processor node or loss of a disk controller does not decrease data availability.
Cliques are the physical medium that supports the migration of vprocs during times of node failure. If a node in a clique fails, then its vprocs migrate to another node in the clique and continue to operate while recovery occurs on their home node.
PEs for channel-attached hardware cannot migrate because they are dependent on the hardware that is physically attached to the node to which they are assigned.
PEs for LAN-attached connections do migrate when a node failure occurs, as do all AMPs.