Interconnection networks for large-scale parallel processing: theory and case studies (2nd ed.)
Interconnection networks for large-scale parallel processing: theory and case studies (2nd ed.)
The network architecture of the Connection Machine CM-5 (extended abstract)
SPAA '92 Proceedings of the fourth annual ACM symposium on Parallel algorithms and architectures
Processor allocation and fault tolerance in multiprocessors
Processor allocation and fault tolerance in multiprocessors
A Distributed Resource Management Mechanism for a Partitionable Multiprocessor System
IEEE Transactions on Computers
The Theory Underlying the Partitioning of Permutation Networks
IEEE Transactions on Computers
On a Class of Multistage Interconnection Networks
IEEE Transactions on Computers
Hi-index | 0.00 |
Partitionability allows the creation of many physically independent subsystems, each of which retains an identical functionality as its parent network and has no communication interference with other subsystems. We show that different permutation functions connecting the processors and the switches in the last stage of networks result in different partitionability. Based on a novel mapping scheme of MINs onto the hypercube structure, we show that the switches play a more important role on the subsystem availability than the processors. Subsystem fault tolerance of this class of MINs is also analyzed.