Journal of Computer Applications ›› 2018, Vol. 38 ›› Issue (3): 707-714.DOI: 10.11772/j.issn.1001-9081.2017092311

Previous Articles     Next Articles

Cloud task scheduling strategy based on clustering and improved symbiotic organisms search algorithm

LI Kunlun, GUAN Liwei, GUO Changlong   

  1. Electronic Information Engineering College, Hebei University, Baoding Heibei 071000, China
  • Received:2017-09-26 Revised:2017-11-16 Online:2018-03-10 Published:2018-03-07
  • Supported by:
    This work is partially supported by the National Natural Science Foundation of China (61672205).

基于聚类和改进共生演算法的云任务调度策略

李昆仑, 关立伟, 郭昌隆   

  1. 河北大学 电子信息工程学院, 河北 保定 071000
  • 通讯作者: 李昆仑
  • 作者简介:李昆仑(1962-),男,河北保定人,教授,博士,CCF会员,主要研究方向:模式识别、图像处理、计算机网络、智能信息处理;关立伟(1990-),男(满族),河北承德人,硕士研究生,主要研究方向:云计算任务调度、虚拟化;郭昌隆(1991-),男,河北石家庄人,硕士研究生,主要研究方向:数据挖掘、推荐算法。
  • 基金资助:
    国家自然科学基金资助项目(61672205)。

Abstract: To solve the problems of some Quality of Service (QoS)-based scheduling algorithms in cloud computing environment, such as slow optimizing speed and imbalance between scheduling cost and user satisfaction, a cloud task scheduling strategy based on clustering and improved SOS (Symbiotic Organisms Search) algorithm was proposed. Firstly, the tasks and resources were clustered by fuzzy clustering and the resources were reordered and placed, and then the tasks were guided and assigned according to the similarity of attributes to reduce the selection range of resources. Secondly, the SOS algorithm was improved according to the cross and rotation learning mechanism to improve the algorithm search ability. Finally, the driving model was constructed by weighted summation to balance the relationship between scheduling cost and system performance. Compared with the improved global genetic algorithm, hybrid particle swarm optimization and genetic algorithm, and discrete SOS algorithm, the proposed algorithm can effectively reduce the evolution generation, reduce the scheduling cost and improve the user's satisfaction. Experimental results show that the proposed algorithm is a feasible and effective task scheduling algorithm.

Key words: cloud computing, Quality of Service (QoS), fuzzy clustering, Symbiotic Organisms Search (SOS) algorithm, task scheduling

摘要: 针对云计算环境中一些基于服务质量(QoS)调度算法存在寻优速度慢、调度成本与用户满意度不均衡的问题,提出了一种基于聚类和改进共生演算法的云任务调度策略。首先将任务和资源进行模糊聚类并对资源进行重排序放置,依据属性相似度对任务进行指导分配,减小对资源的选择范围;然后依据交叉和旋转学习机制改进共生演算法,提升算法的搜索能力;最后通过加权求和方式构造驱动模型,均衡调度代价与系统性能间关系。通过不同任务量的云任务调度仿真实验,表明该算法相比改进遗传算法、混合粒子群遗传算法和离散共生演算法,有效减少了进化代数,降低了调度成本并提升了用户满意度,是一种可行有效的任务调度算法。

关键词: 云计算, 服务质量, 模糊聚类, 共生演算法, 任务调度

CLC Number: