Adaptive Task Migration Scheme in Distributed Imprecise Computing

  • RUSDIUTOMO Amien
    Graduate School of Information Systems, University of Electro-Communications
  • SATO Naohito
    Graduate School of Information Systems, University of Electro-Communications
  • HONDA Hiroki
    Graduate School of Information Systems, University of Electro-Communications
  • YUBA Toshitugu
    Graduate School of Information Systems, University of Electro-Communications

Bibliographic Information

Other Title
  • 分散インプリサイス計算における負荷の状態近似に基づく適応的なタスク移送方式

Search this article

Description

In distributed imprecise computing, some of computing nodes may often become overloaded due to uneven task arrivals while the other nodes are left under loaded. Consequently, some tasks may miss their deadlines to complete their mandatory parts, or fail to process their optional parts even when the distributed system has enough processing resources. In this paper, we propose a new symmetrically initiated scheduling scheme based on load approximation to increase the number of processing tasks. Our scheduling scheme exploits state information of the nodes to combine sender initiated scheme and receiver initiated scheme.

Journal

References(7)*help

See more

Details 詳細情報について

  • CRID
    1573387452198944384
  • NII Article ID
    110003180343
  • NII Book ID
    AN10013141
  • Text Lang
    ja
  • Data Source
    • CiNii Articles

Report a problem

Back to top