• Title/Summary/Keyword: 적응적 부하재분배 알고리즘

Search Result 2, Processing Time 0.028 seconds

A GA-Based Adaptive Task Redistribution Method for Intelligent Distributed Computing (지능형 분산컴퓨팅을 위한 유전알고리즘 기반의 적응적 부하재분배 방법)

  • 이동우;이성훈;황종선
    • Journal of KIISE:Software and Applications
    • /
    • v.31 no.10
    • /
    • pp.1345-1355
    • /
    • 2004
  • In a sender-initiated load redistribution algorithm, a sender(overloaded processor) continues to send unnecessary request messages for load transfer until a receiver(underloaded processor) is found while the system load is heavy. In a receiver-initiated load redistribution algorithm, a receiver continues to send unnecessary request messages for load acquisition until a sender is found while the system load is light. Therefore, it yields many problems such as low CPU utilization and system throughput because of inefficient inter-processor communications in this environment. This paper presents an approach based on genetic algorithm(GA) for adaptive load sharing in distributed systems. In this scheme, the processors to which the requests are sent off are determined by the proposed GA to decrease unnecessary request messages.

A Genetic-based Methodology for Adjustable Load Redistribution (적응성 있는 부하 재분배를 위한 유전적 방법론)

  • Lee, Seong-Hoon
    • Proceedings of the Korean Information Science Society Conference
    • /
    • 2005.11b
    • /
    • pp.691-693
    • /
    • 2005
  • In a sender-initiated load balancing algorithm, a sender continues to send unnecessary request messages for load transfer until a receiver is found while the system load is heavy. Therefore, it yields many problems such as low cpu utilization and system throughput because of inefficient inter-processor communications until the sender receives an accept message from the receiver in this environment. This paper presents an approach based on evolutionary algorithm. The processors to which the requests are sent off are determined by the proposed algorithm to decrease unnecessary request messages.

  • PDF