Adaptive Fault-Tolerant Deadlock-Free Routing in Meshes and Hypercubes
IEEE Transactions on Computers
An Improved Algorithm for Fault-Tolerant Wormhole Routing in Meshes
IEEE Transactions on Computers
A Fault-Tolerant Routing Scheme for Meshes with Nonconvex Faults
IEEE Transactions on Parallel and Distributed Systems
Interconnection Networks: An Engineering Approach
Interconnection Networks: An Engineering Approach
Fault-Tolerant Wormhole Routing Algorithms for Mesh Networks
IEEE Transactions on Computers
Communication in Multicomputers with Nonconvex Faults
IEEE Transactions on Computers
A New Approach to Fault-Tolerant Wormhole Routing for Mesh-Connected Parallel Computers
IPDPS '02 Proceedings of the 16th International Parallel and Distributed Processing Symposium
Origin-based fault-tolerant routing in the mesh
HPCA '95 Proceedings of the 1st IEEE Symposium on High-Performance Computer Architecture
Fault-Tolerant Wormhole Routing Algorithms in Meshes in the Presence of Concave Faults
IPDPS '00 Proceedings of the 14th International Symposium on Parallel and Distributed Processing
Blue Gene: a vision for protein science using a petaflop supercomputer
IBM Systems Journal - Deep computing for the life sciences
Hi-index | 0.00 |
A fault-tolerant routing method that can tolerate solid faults using only two virtual channels is presented. The proposed routing algorithm, called FT-Ecube, not only uses a fewer number of virtual channels but also tolerates f-chains in the meshes. Furthermore, the proposed scheme misroutes messages both clockwise and counter clockwise directions to reduce channel contention on f-rings. It is shown that the proposed algorithm is deadlock-free and livelock-free in meshes when it has nonoverlapping multiple f-regions. Further, we conducted flit-level simulations to evaluate the performance of the proposed routing algorithm. As our simulation results show, FT-Ecube tolerates multiple faulty blocks using only two virtual channels per physical channel, and has good performance in terms of average latency.