Task Assignment in Distributed Systems based on PSO Approach

In a distributed system, Task Assignment Problem (TAP) is a key factor for obtaining efficiency. TAP illustrates the appropriate allocation of tasks to the processor of each computer. In this problem, the proposed methods up to now try to minimize Makespan and maximizing CPU utilization. Since this problem is NP-complete, many genetic algorithms have been proposed to search optimal solutions from the entire solution space. Disregarding the techniques which can reduce the complexity of optimization, the existing approaches scan the entire solution space. On the other hand, this approach is time-consuming in scheduling which is considered a shortcoming. Therefore, in this paper, a hybrid genetic algorithm has been proposed to overcome this shortcoming. Particle Swarm Optimization (PSO) has been applied as local search in the proposed genetic algorithm in this paper. The results obtained from simulation can prove that, in terms of CPU utilization and Makespan, the proposed approach outperforms the GA-based approach.

[1]  Mahdi Jameii,et al.  A new distributed systems scheduling algorithm: a swarm intelligence approach , 2011, Other Conferences.

[2]  B. Freisleben,et al.  A comparison of memetic algorithms, tabu search, and ant colonies for the quadratic assignment problem , 1999, Proceedings of the 1999 Congress on Evolutionary Computation-CEC99 (Cat. No. 99TH8406).

[3]  M. H. Kashani,et al.  PARALLEL MIN-MAX ANT COLONY SYSTEM ( MMAS ) FOR DYNAMIC PROCESS SCHEDULING IN DISTRIBUTED OPERATING SYSTEMS CONSIDERING LOAD BALANCING , 2007 .

[4]  Walter H. Kohler,et al.  Models for Dynamic Load Balancing in a Heterogeneous Multiple Processor System , 1979, IEEE Transactions on Computers.

[5]  Raheleh Sarvizadeh,et al.  A swarm intelligence based memetic algorithm for task allocation in distributed systems , 2012, International Conference on Machine Vision.

[6]  Masahiro Tsuchiya,et al.  A Task Allocation Model for Distributed Computing Systems , 1982, IEEE Transactions on Computers.

[7]  Hisao Ishibuchi,et al.  Balance between genetic search and local search in memetic algorithms for multiobjective permutation flowshop scheduling , 2003, IEEE Trans. Evol. Comput..

[8]  M. H. Kashani,et al.  A GENETIC ALGORITHM FOR PROCESS SCHEDULING IN DISTRIBUTED OPERATING SYSTEMS CONSIDERING LOAD BALANCING , 2007 .

[9]  A. J. Urdaneta,et al.  A Hybrid Particle Swarm Optimization for Distribution State Estimation , 2002, IEEE Power Engineering Review.

[10]  M. H. Kashani,et al.  A New Method Based on Memetic Algorithm for Task Scheduling in Distributed Systems , 2010 .

[11]  Jinyuan You,et al.  Main sequences genetic scheduling for multiprocessor systems using task duplication , 2004, Microprocess. Microsystems.

[12]  Riccardo Poli,et al.  Particle swarm optimization , 1995, Swarm Intelligence.

[13]  Chien-Chung Shen,et al.  A Graph Matching Approach to Optimal Task Assignment in Distributed Computing Systems Using a Minimax Criterion , 1985, IEEE Transactions on Computers.

[14]  Mohsen Jahanshahi,et al.  Using Simulated Annealing for Task Scheduling in Distributed Systems , 2009, 2009 International Conference on Computational Intelligence, Modelling and Simulation.

[15]  M. H. Kashani,et al.  A novel method for task scheduling in distributed systems using Max-Min Ant Colony Optimization , 2011, 2011 3rd International Conference on Advanced Computer Control.

[16]  Peng-Yeng Yin,et al.  A hybrid particle swarm optimization algorithm for optimal task assignment in distributed systems , 2006, Comput. Stand. Interfaces.

[17]  Yue Shi,et al.  A modified particle swarm optimizer , 1998, 1998 IEEE International Conference on Evolutionary Computation Proceedings. IEEE World Congress on Computational Intelligence (Cat. No.98TH8360).

[18]  Mostafa Haghi Kashani,et al.  A Novel Bee Colony Approach to Distributed Systems Scheduling , 2012 .

[19]  M. Akbari,et al.  Utilizing Bee Colony to Solve Task Scheduling Problem in Distributed Systems , 2011, 2011 Third International Conference on Computational Intelligence, Communication Systems and Networks.