• DocumentCode
    2205040
  • Title

    A new nature-inspired algorithm for load balancing

  • Author

    Feng, Xiang ; Lau, Francis C M ; Shuai, Dianxun

  • Author_Institution
    Dept. of Comput. Sci., East China Univ. of Sci. & Technol., Shanghai, China
  • fYear
    2008
  • fDate
    19-21 Nov. 2008
  • Firstpage
    289
  • Lastpage
    293
  • Abstract
    The classical Load Balancing Problem (LBP) is to map tasks to processors so as to minimize the maximum load. Solving the LBP successfully would lead to better utilization of resources and better performance. The LBP has been proven to be NP-hard, thus generating the exact solutions in a tractable amount of time becomes infeasible when the problems become large.We present a new nature-inspired approximation algorithm based on the Particle Mechanics (PM) model to compute in parallel approximate efficient solutions for LBPs. Just like other Nature-inspired Algorithms (NAs) drawing from observations of physical processes that occur in nature, the PM algorithm is inspired by physical models of particle kinematics and dynamics. The PM algorithm maps the classical LBP to the movement of particles in a force field by a corresponding mathematical model in which all particles move according to certain defined rules until reaching a stable state. By anti-mapping the stable state, the solution to LBP can be obtained.
  • Keywords
    Lyapunov matrix equations; parallel machines; resource allocation; approximation algorithm; distributed algorithm; load balancing problem; nature-inspired algorithms; parallel algorithm; particle mechanics; Approximation algorithms; Computer science; Concurrent computing; Heuristic algorithms; Kinematics; Load management; Mathematical model; Parallel algorithms; Parallel machines; Spine; Load balancing; approximation algorithm; distributed and parallel algorithm; nature-inspired algorithm; particle mechanics model;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Communication Systems, 2008. ICCS 2008. 11th IEEE Singapore International Conference on
  • Conference_Location
    Guangzhou
  • Print_ISBN
    978-1-4244-2423-8
  • Electronic_ISBN
    978-1-4244-2424-5
  • Type

    conf

  • DOI
    10.1109/ICCS.2008.4737190
  • Filename
    4737190