Advanced Hybrid Information Processing. First International Conference, ADHIP 2017, Harbin, China, July 17–18, 2017, Proceedings

Research Article

An Improved Genetic Algorithm on Task Scheduling

Download
173 downloads
  • @INPROCEEDINGS{10.1007/978-3-319-73317-3_57,
        author={Fangyuan Zheng and Jingmei Li},
        title={An Improved Genetic Algorithm on Task Scheduling},
        proceedings={Advanced Hybrid Information Processing. First International Conference, ADHIP 2017, Harbin, China, July 17--18, 2017, Proceedings},
        proceedings_a={ADHIP},
        year={2018},
        month={2},
        keywords={Task scheduling Heterogeneous multi-core processor Genetic algorithm Optimal solution},
        doi={10.1007/978-3-319-73317-3_57}
    }
    
  • Fangyuan Zheng
    Jingmei Li
    Year: 2018
    An Improved Genetic Algorithm on Task Scheduling
    ADHIP
    Springer
    DOI: 10.1007/978-3-319-73317-3_57
Fangyuan Zheng1,*, Jingmei Li1
  • 1: Harbin Engineering University
*Contact email: zhengfangyuan@hrbeu.edu.cn

Abstract

Efficient task scheduling algorithm is critical for achieving high performance in heterogeneous multi-core processors. Because the existing genetic algorithm converges to local optimal solution, so an improved genetic algorithm is proposed to solve the above problems in this thesis. Firstly, the initial population is generated randomly according to the task height value, and then adopting the selection strategy based on competition scale. Finally, the crossover and mutation probability is improved to avoid premature phenomenon. The experiment based on randomly generated graphs shows that the proposed algorithm can improve the efficiency of convergence.