Real-time fault-tolerant hypercube multicomputer

A real-time fault-tolerant design for a d-dimensional hypercube multiprocessor with two modes of operation is presented and its reconfigurability is examined. The augmented hypercube, at stage one, has a spare node connected to each node of a subcube of dimension i, and the spare nodes are also connected as a (d-i)-dimensional hypercube. At stage two, the process is repeated by assigning one spare node to each (d-i-j)-dimensional spare subcube of stage one. Two modes of operations are considered, one under heavy computation or hard deadline and the other under light computation or soft deadline. By utilising the capabilities of wave-switching communication modules of the spare nodes, faulty nodes and faulty links can be tolerated. Both theoretical and experimental results are presented. Compared with other proposed schemes, the proposed approach can tolerate significantly more faulty components with a low overhead and no performance degradation.