The connection machine
Deadlock-Free Message Routing in Multiprocessor Interconnection Networks
IEEE Transactions on Computers
The art of computer programming, volume 3: (2nd ed.) sorting and searching
The art of computer programming, volume 3: (2nd ed.) sorting and searching
ACM Transactions on Programming Languages and Systems (TOPLAS)
Routing, merging and sorting on parallel models of computation
STOC '82 Proceedings of the fourteenth annual ACM symposium on Theory of computing
Universal schemes for parallel communication
STOC '81 Proceedings of the thirteenth annual ACM symposium on Theory of computing
STOC '83 Proceedings of the fifteenth annual ACM symposium on Theory of computing
New bounds for parallel prefix circuits
STOC '83 Proceedings of the fifteenth annual ACM symposium on Theory of computing
Computational Aspects of VLSI
Partitioning Message Patterns for Bundled Omega Networks
IEEE Transactions on Parallel and Distributed Systems
A journey into multicomputer routing algorithms
PAS '95 Proceedings of the First Aizu International Symposium on Parallel Algorithms/Architecture Synthesis
Hi-index | 14.98 |
A deterministic routing scheme for a communications network based on the k-dimensional hypercube is proposed. The author presents two formulations of the scheme. The first formulation delivers messages in O(k/sup 2/) bit times using O(K) bits of buffer space at each node in the hypercube. The second formulation assumes that there are several batches of messages to be delivered, and makes certain assumptions about the cost of sending messages along the various dimensions of the cube. In this case, the latency for delivery time is still O(k/sup 2/) bit times, but the throughput is increased to one set of messages every O(k) bit times. For the first formulation, the author restricts himself to routings which are subsets of permutations (i.e. every node sends at most one message and receives at most one message). The second formulation indicates a way to perform routings which are subsets of H-permutations (i.e. every node sends at most H messages and receives at most H messages).