Scheduling in Heterogeneous Distributed Systems

Parallel and distributed systems play an important part in the improvement of high-performance computing. In analyzing the performance of such Heterogeneous Distributed Systems (HDS), scheduling a set of tasks to the available set of resources for execution is highly important. Task-scheduling being an NP-complete problem, use of meta-heuristics is more appropriate in obtaining optimal solutions. The problem of scheduling an application comprising a set of independent tasks on a fully connected HDS is considered here. The scheduling algorithms for this problem focus on minimizing two objectives, the make-span and the flow-time. These objectives conflict with one another, which requires multi-objective problem formulation. Multi-objective Genetic Algorithms (MOGAs) and Multi-objective Particle Swarm Optimization (MOPSO) algorithms are applied to the problem. Weighted Sum Genetic Algorithm (WSGA) based on weighted sum approach, and Nondominated Sorting Genetic Algorithm (NSGA2), a modified version of NSGA2 with controlled elitism (NSGA2-CE) and a hybrid version of NSGA2 combined with Pareto hill climbing (PHC), the NSGA2-PHC are applied to the problem. A weighted sum particle swarm optimization (WSPSO), a Nondominated Sorting Particle Swarm Optimization (NSPSO), an Adaptive Nondominated Sorting Particle Swarm Optimization (ANSPSO), and a hybrid version of ANSPSO with PHC, the NSPSO-PHC are the MOPSO methods applied to task scheduling. The different MOGA and MOPSO methods are compared among themselves to determine the algorithms that generate the efficient schedule. The best version of MOGA is compared with the best MOPSO method to find that MOGA produces better schedules for benchmark test instances simulated.

[1]  Edward G. Coffman,et al.  Computer and job-shop scheduling theory , 1976 .

[2]  Lothar Thiele,et al.  Comparison of Multiobjective Evolutionary Algorithms: Empirical Results , 2000, Evolutionary Computation.

[3]  Hongbo Liu,et al.  Nature inspired meta-heuristics for grid scheduling: single and multi-objective optimization approaches , 2008 .

[4]  Kalyanmoy Deb,et al.  Multi-objective optimization using evolutionary algorithms , 2001, Wiley-Interscience series in systems and optimization.

[5]  Yuehui Chen,et al.  A Task Scheduling Algorithm Based on PSO for Grid Computing , 2008 .

[6]  M C BHUVANESWARI,et al.  Comparison of multi-objective evolutionary approaches for task scheduling in distributed computing systems , 2012, Sadhana.

[7]  Thomas M. Keane,et al.  Multi-heuristic dynamic task allocation using genetic algorithms in a heterogeneous distributed system , 2010, J. Parallel Distributed Comput..

[8]  Howard Jay Siegel,et al.  Representing Task and Machine Heterogeneities for Heterogeneous Computing Systems , 2000 .

[9]  Peter J. Fleming,et al.  Evolutionary algorithms in control systems engineering: a survey , 2002 .

[10]  David S. Johnson,et al.  Computers and Intractability: A Guide to the Theory of NP-Completeness , 1978 .

[11]  Xiaodong Li,et al.  A Non-dominated Sorting Particle Swarm Optimizer for Multiobjective Optimization , 2003, GECCO.

[12]  A. Abraham,et al.  Scheduling jobs on computational grids using a fuzzy particle swarm optimization algorithm , 2010, Future Gener. Comput. Syst..

[13]  Sajal K. Das,et al.  Optimizing QoS-Based Multicast Routing in Wireless Networks: A Multi-objective Genetic Algorithmic Approach , 2002, NETWORKING.

[14]  Jan Platos,et al.  Differential Evolution for Scheduling Independent Tasks on Heterogeneous Distributed Environments , 2010 .

[15]  Jing He,et al.  A PSO-based Genetic Algorithm for Scheduling of Tasks in a Heterogeneous Distributed System , 2013, J. Softw..

[16]  Fatos Xhafa,et al.  Genetic algorithm based schedulers for grid computing systems , 2007 .

[17]  Ian Foster,et al.  The Grid 2 - Blueprint for a New Computing Infrastructure, Second Edition , 1998, The Grid 2, 2nd Edition.

[18]  Ladislau Bölöni,et al.  A Comparison of Eleven Static Heuristics for Mapping a Class of Independent Tasks onto Heterogeneous Distributed Computing Systems , 2001, J. Parallel Distributed Comput..