The classification of coverings of processor networks
Journal of Parallel and Distributed Computing
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.)
Introduction to parallel algorithms and architectures: array, trees, hypercubes
Introduction to parallel algorithms and architectures: array, trees, hypercubes
Broadband switching: architectures, protocols, design, and analysis
Broadband switching: architectures, protocols, design, and analysis
Time-Division Optical Communications in Multiprocessor Arrays
IEEE Transactions on Computers
Hierarchical Scalable Photonic Architectures for High-Performance Processor Interconnection
IEEE Transactions on Computers
| Hi-index | 0.00 |
A passive optical realization of large expandable shuffle networks is considered, where the general (p, k) shuffle function interconnects star-coupled clusters of time and/or wavelength multiplexed nodes. This configuration enables network partitioning into independent subnetworks that can emulate various indirect cube topologies on a virtual point-to-point basis. Node transmitters are assigned fixed channels and reconfiguring the partition and/or the individual sub-network topologies is achieved by monitoring the appropriate channels. The considered network configuration can function as a general-purpose optical interconnect for a variety of heterogeneous traffic sources, such as multicomputers and ATM network interface units, communicating independently within logically defined subnetworks. The paper derives a conflict-free channel partition and assignment map for a general (p, k) cluster shuffle using a minimal number of channel sets.