Comparison of Heuristics for Scheduling Independent Tasks on Heterogeneous Distributed Environments
Hesam Izakian, Ajith Abraham, Václav Snåšel · 2009
Scheduling is one of the core steps to efficientlyexploit the capabilities of heterogeneous distributedcomputing systems and is an NP-complete problem.Therefore using meta-heuristic algorithms is asuitable approach in order to cope with its difficulty.In meta-heuristic algorithms, generating individualsin the initial step has an important effect on theconvergence behavior of the algorithm and finalsolutions. Using some heuristics for generating one ormore near-optimal individuals in the initial step canimprove the final solutions obtained by meta-heuristicalgorithms. Different criteria can be used forevaluating the efficiency of scheduling algorithms, themost important of which are makespan and flowtime.In this paper we propose an efficient heuristic methodand then we will compare with five popular heuristicsfor minimizing makespan and flowtime inheterogeneous distributed computing systems.