ENHANCED HYBRID PSO – ACO ALGORITHM FOR GRID SCHEDULING
Autor: | Sivanandam S N, Mathiyalagan P, Dhepthie U R |
---|---|
Jazyk: | angličtina |
Rok vydání: | 2010 |
Předmět: |
Grid Scheduling
Mathematical optimization Grid scheduling lcsh:Computer engineering. Computer hardware Swarm Intelligence Inertia Computer science Computer Science::Neural and Evolutionary Computation MathematicsofComputing_NUMERICALANALYSIS Pheromone lcsh:TK7885-7895 ComputingMethodologies_ARTIFICIALINTELLIGENCE |
Zdroj: | ICTACT Journal on Soft Computing, Vol 1, Iss 1, Pp 54-59 (2010) |
ISSN: | 2229-6956 0976-6561 |
Popis: | Grid computing is a high performance computing environment to solve larger scale computational demands. Grid computing contains resource management, task scheduling, security problems, information management and so on. Task scheduling is a fundamental issue in achieving high performance in grid computing systems. A computational GRID is typically heterogeneous in the sense that it combines clusters of varying sizes, and different clusters typically contains processing elements with different level of performance. In this, heuristic approaches based on particle swarm optimization and ant colony optimization algorithms are adopted for solving task scheduling problems in grid environment. Particle Swarm Optimization (PSO) is one of the latest evolutionary optimization techniques by nature. It has the better ability of global searching and has been successfully applied to many areas such as, neural network training etc. Due to the linear decreasing of inertia weight in PSO the convergence rate becomes faster, which leads to the minimal makespan time when used for scheduling. To make the convergence rate faster, the PSO algorithm is improved by modifying the inertia parameter, such that it produces better performance and gives an optimized result. The ACO algorithm is improved by modifying the pheromone updating rule. ACO algorithm is hybridized with PSO algorithm for efficient result and better convergence in PSO algorithm. |
Databáze: | OpenAIRE |
Externí odkaz: |