Nodes and disk arrays are interconnected into a clique configuration for fault tolerance. If a node in a clique fails, its tasks are automatically moved over to the other nodes in the clique. The other nodes in the clique have direct access to all disk arrays in the clique. The system continues running minus the failed node.
Each Teradata database processing (TPA) node has redundant connections to the disk array in its clique. All links are point to point.
Node to Disk Array Interconnect: 4+0 Production Clique (One Quad 6 Gbx4 SAS HBA)
Node to Disk Array Interconnect: 2+0 2800 Production Clique or Second 2800 System (One Quad 6 Gbx4 SAS HBA)
Node to Disk Array Interconnect: 680 System (One Quad 6 Gbx4 SAS HBA)