Short Notes A Distributed Management Scheme for Partitionable Parallel Computers

Academic Article

Abstract

  • Many large-scale parallel computers, such as those based on hypercube or multistage cube interconnection networks, can be partitioned to process applications with different computation structures and different degrees of parallelism simultaneously. However, partitioning may create resource fragments and may result in a loss of computation power. Dynamic partitioning is an effective method to alleviate the fragmentation problem. A distributed scheme for dynamic partitioning is investigated in this paper. Distributed procedures to split a subsystem and to combine subsystems are presented. Correctness of each of these two procedures is shown and the complexity is analyzed. The procedures presented are applicable to parallel computers that use interconnection networks such as hypercube, omega, multistage cube, and extra stage cube networks. © 1990 IEEE
  • Authors

    Digital Object Identifier (doi)

    Author List

  • Jeng M; Siegel HJ
  • Start Page

  • 120
  • End Page

  • 126
  • Volume

  • 1
  • Issue

  • 1