A Permutation-Based Differential Evolution Algorithm Incorporating Simulated Annealing for Multiprocessor Scheduling with Communication Delays

  • Xiaohong Kong
  • Wenbo Xu
  • Jing Liu
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3991)


Employing a differential evolution (DE) algorithm, we present a novel permutation-based search technique in list scheduling for parallel program. By encoding a vector as a scheduling list and differential variation as s swap operator, the DE algorithm can generate high quality solutions in a short time. In standard differential evolution algorithm, while constructing the next generation, a greedy strategy is used which maybe lead to convergence to a local optimum. In order to avoid the above problem, we combine differential evolution algorithm with simulated annealing algorithm which relaxes the criterion selecting the next generation. We also use stochastic topological sorting algorithm (STS) to generate an initial scheduling list. The results demonstrate that the hybrid differential evolution generates better solutions even optimal solutions in most cases and simultaneously meet scalability.


Simulated Annealing Directed Acyclic Graph Task Graph Communication Delay Schedule List 


  1. 1.
    Ahmad, I., Dhodhi, M.K.: Multiprocessor Scheduling in a Genetic Paradigm. Parallel Computing 22, 395–406 (1996)MATHCrossRefGoogle Scholar
  2. 2.
    Kwok, Y.-K., Ahmad, I.: Efficient Scheduling of Arbitrary Task Graphs to Multiprocessors Using a Parallel Genetic Algorithm. J. Parallel and Distributed Computing 47, 58–77 (1997)CrossRefGoogle Scholar
  3. 3.
    Davidovic, T., Hansen, P., Mladenovic, N.: Permutation Based Genetic, Tabu and Variable Neighborhood Search Heuristics for Multiprocessor Scheduling with Communication Delays. Asia-Pacific Journal of Operational Research 22, 297–326 (2005)MATHCrossRefMathSciNetGoogle Scholar
  4. 4.
    Davidovic, T., Crainic, T.G.: New Benchmarks for Static Task Scheduling on Homogeneous Multiprocessor Systems with Communication Delays. Publication CRT-2003-04, Centre de Recherche sur les Transports, Universite de Montreal (2003)Google Scholar
  5. 5.
    Davidovic, T., Hansen, P., Mladenovic, N.: Variable Neighborhood Search for Multiprocessor Scheduling Problem with Communication Delays. In: de Sous, J.P. (ed.) Proc. MIC 2001, 4th Metaheuristic International Conference, Porto, Portugal (2001)Google Scholar
  6. 6.
    Storn, R., Price, K.: Differential Evolution-a Simple and Efficient Heuristic for Global Optimization over Continuous Space. Journal of Global Optimization 11, 341–359 (1997)MATHCrossRefMathSciNetGoogle Scholar
  7. 7.
    Storn, R.: System Design by Constraint Adaptation and Differential Evolution. IEEE Transaction on Evolutionary Computation 3, 22–34 (1999)CrossRefGoogle Scholar
  8. 8.
    Wang, K.-P., Huang, L., Zhou, C.-G., Pang, W.: Particle Swarm Optimization for Traveling Salesman Problem. In: Proceedings of the Second International Conference on Machine Learning and Cybernetics, vol. 5, pp. 1583–1585 (2003)Google Scholar
  9. 9.
    Xu, W., Sun, J.: Efficient Scheduling of Task Graphs to Multiprocessors Using a Simulated Annealing Algorithm. In: DCABES 2004 Proceedings, vol. 1, pp. 435–439 (2004)Google Scholar
  10. 10.
    Goldberg, D.E.: Genetic Algorithms in Search, Optimization, and Matching Learning. Addison-Wesley Publishing Company, Inc., Reading (1989)Google Scholar
  11. 11.
    Zhang, H., Li, X., Li, H.: Particle Swarm Optimization-Based Schemes for Resource-Constrained Project Scheduling. Automation in Construction 14, 393–404 (2005)CrossRefGoogle Scholar
  12. 12.
    Kirkpatrick, S., Gelatt, C., Vecchi, M.: Optimization by Simulated Annealing. Science 220, 671–680 (1983)CrossRefMathSciNetGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Xiaohong Kong
    • 1
    • 2
  • Wenbo Xu
    • 1
  • Jing Liu
    • 1
  1. 1.School of Information TechnologySouthern Yangtze UniversityWuxiChina
  2. 2.Henan Institute Of Science and TechnologyXinxiangChina

Personalised recommendations