Fault-tolerant meshes with efficient layouts

This paper presents a practical fault-tolerant architecture for mesh parallel machines that has only one spare processor and has only six communication links per processor while tolerating one processor fault and one communication link fault, or two communication link faults. We also show that the architecture presented here can be laid out efficiently in a linear area with wire length at most six.

[1]  Jehoshua Bruck,et al.  Fault-Tolerant Meshes and Hypercubes with Minimal Numbers of Spares , 1993, IEEE Trans. Computers.

[2]  Koji Yamamoto,et al.  Fault-tolerant graphs for hypercubes and tori , 1995, Proceedings of the Twenty-Eighth Annual Hawaii International Conference on System Sciences.

[3]  Jehoshua Bruck,et al.  Fault-tolerant meshes with minimal numbers of spares , 1991, Proceedings of the Third IEEE Symposium on Parallel and Distributed Processing.

[4]  Jehoshua Bruck,et al.  Tolerating Faults in a Mesh with a Row of Spare Nodes , 1994, Theor. Comput. Sci..

[5]  John P. Hayes,et al.  A Graph Model for Fault-Tolerant Computing Systems , 1976, IEEE Transactions on Computers.

[6]  John P. Hayes,et al.  Designing Fault-Tolerant System Using Automorphisms , 1991, J. Parallel Distributed Comput..

[7]  John P. Hayes,et al.  Edge fault tolerance in graphs , 1993, Networks.

[8]  Jehoshua Bruck,et al.  Wildcard dimensions, coding theory and fault-tolerant meshes and hypercubes , 1993, FTCS-23 The Twenty-Third International Symposium on Fault-Tolerant Computing.

[9]  Jehoshua Bruck,et al.  Fault-tolerant meshes with small degree , 1993, SPAA '93.