Tight bounds on the simulation of meshes with dynamically reconfigurable row/column buses by meshes with statically partitioned buses

Authors:
Susumu Matsumae
Affiliations:
Department of Information System, Tottori University of Environmental Studies, Tottori-City, Tottori, Japan
Venue:
Journal of Parallel and Distributed Computing
Year:
2006

Citing 9
Cited 0

Array processor with multiple broadcasting

Journal of Parallel and Distributed Computing
Meshes with reconfigurable buses

Proceedings of the fifth MIT conference on Advanced research in VLSI
Prefix Computations on a Generalized Mesh-Connected Computer with Multiple Buses

IEEE Transactions on Parallel and Distributed Systems
Efficient self-simulation algorithms for reconfigurable arrays

Journal of Parallel and Distributed Computing
An Improved Generalization of Mesh-Connected Computers with Multiple Buses

IEEE Transactions on Parallel and Distributed Systems
Using bus linearization to scale the reconfigurable mesh

Journal of Parallel and Distributed Computing
Constant Time Algorithms for the Transitive Closure and Some Related Graph Problems on Processor Arrays with Reconfigurable Bus Systems

IEEE Transactions on Parallel and Distributed Systems
Optimal Architectures and Algorithms for Mesh-Connected Parallel Computers with Separable Row/Column Buses

IEEE Transactions on Parallel and Distributed Systems
Sourcebook of parallel computing

Sourcebook of parallel computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper studies the simulation problem of meshes with separable buses (MSB) by meshes with multiple partitioned buses (MMPB). The MSB and the MMPB are the mesh-connected computers with row/column broadcasting buses. The broadcasting buses of the MSB can be dynamically sectioned into smaller bus segments by program control, while those of the MMPB are statically partitioned in advance. In the MSB, each row/column has only one separable bus, while in the MMPB, each row/column has L partitioned buses (L≥ 1, L is a fixed constant). Two results are shown: (1) the n × n MMPB can simulate any step of the n × n MSB time-optimally in O(n1/(2L+1)) steps, and (2) the m × m MMPB with L = 1 can simulate any step of the n × n MSB in O(n2/m2 + n1/3) steps (m ≤ n). The latter implies that the MMPB with L - 1 can solve the scaling-simulation problem of the MSB time-optimally when m = O(n5/6) holds.