Performance Analysis of k-ary n-cube Interconnection Networks
IEEE Transactions on Computers
Processor resource management in partitionable parallel computers
Processor resource management in partitionable parallel computers
Optimal and Suboptimal Processor Allocation for Hypercycle-based Multiprocessors
IEEE Transactions on Parallel and Distributed Systems
Advanced Computer Architecture: Parallelism,Scalability,Programmability
Advanced Computer Architecture: Parallelism,Scalability,Programmability
ProcSimity: an experimental tool for processor allocation and scheduling in highly parallel systems
FRONTIERS '95 Proceedings of the Fifth Symposium on the Frontiers of Massively Parallel Computation (Frontiers'95)
Distributed dynamic processor allocation for multicomputers
Parallel Computing
Routing-contained virtualization based on Up*/Down* forwarding
HiPC'07 Proceedings of the 14th international conference on High performance computing
Fast and efficient processor allocation algorithm for torus-based chip multiprocessors
Computers and Electrical Engineering
| Hi-index | 0.02 |
Composed of various topologies, the k-ary n-cube system is desirable for accepting and executing topologically different tasks. To utilize its large amount of processor resources, several allocation strategies have been reported, each with certain restrictions that affect performance. For improvement, we propose a new allocation strategy for the k-ary n-cubes. The proposed strategy is an extension of the TC strategy for hypercubes and is able to recognize all subcubes with different topologies requested by tasks. Complexity analysis and performance comparison between related strategies are provided to demonstrate their advantages and disadvantages. Simulation results show that with full subcube recognition ability and no internal fragmentation, our strategy always exhibits better performance.