Flip-Trees: Fault-Tolerant Graphs with Wide Containers
IEEE Transactions on Computers - Fault-Tolerant Computing
Bisectional Fault-Tolerant Communication Architecture for Supercomputer Systems
IEEE Transactions on Computers
The cube-connected cycles: a versatile network for parallel computation
Communications of the ACM
Distributed fault-tolerance for large multiprocessor systems
ISCA '80 Proceedings of the 7th annual symposium on Computer Architecture
Semi-Distributed Load Balancing for Massively Parallel Multicomputer Systems
IEEE Transactions on Software Engineering
Embedding of Cycles in Arrangement Graphs
IEEE Transactions on Computers
Combinatorics, Probability and Computing
Embedding hypercubes, rings, and odd graphs into hyper-stars
International Journal of Computer Mathematics
Embedding algorithms for bubble-sort, macro-star, and transposition graphs
NPC'10 Proceedings of the 2010 IFIP international conference on Network and parallel computing
Optimal Independent Spanning Trees on Odd Graphs
The Journal of Supercomputing
Fault tolerance logical network properties of irregular graphs
ICA3PP'12 Proceedings of the 12th international conference on Algorithms and Architectures for Parallel Processing - Volume Part I
Hi-index | 14.98 |
Odd graphs are analyzed to determine their suitable in designing interconnection networks. These networks are shown to possess many features that make them competitive with other architectures, such as ring, star, mesh, the binary n-cube and its generalized form, the chordal ring, and flip-trees. Among the features are small internode distances, a lighter density, simplicity in implementing various self-routing algorithms (both for faulty and nonfaulty networks), capability of maximal fault tolerance, strong resilience, and good persistence. The routing algorithms (both for the faulty and fault-free networks) do not require any table lookup mechanism, and intermediate nodes do not need to modify the message. These graphs are shown to have a partitioning property that is based on Hadamard matrices and can be effectively used for a system's expansion and self-diagnostics.