计算机集成制造系统 ›› 2019, Vol. 25 ›› Issue (第4): 909-919.DOI: 10.13196/j.cims.2019.04.013

• 当期目次 • 上一篇    下一篇

混合云环境下面向时延优化的科学工作流数据布局策略

林兵1,2,项滔2,3,陈国龙2,3,陈星2,3+   

  1. 1.福建师范大学物理与能源学院
    2.福建省网络计算与智能信息处理重点实验室
    3.福州大学数学与计算机科学学院
  • 出版日期:2019-04-30 发布日期:2019-04-30
  • 基金资助:
    国家重点研发计划资助项目(2017YFB1002000);国家自然科学基金资助项目(61672159);福建省自然科学基金项目(2019J01061386)。

Time-driven data placement strategy for scientific workflows in hybrid cloud

  • Online:2019-04-30 Published:2019-04-30
  • Supported by:
    Project supported by the National Key R&D Program,China(No.2017YFB1002000),the National Natural Science Foundation,China(No.61672159),and the Natural Science Foundation of Fujian Province,China(No.2019J01061386).

摘要: 如何结合公有云和私有云各自的优势,对包含隐私数据的科学工作流数据进行合理布局,优化大规模数据的传输时延,是混合云环境下科学工作流面临的重大挑战。考虑混合云环境下数据布局特点,结合科学工作流数据间的依赖关系,提出一种基于遗传算法算子的自适应离散粒子群优化算法,优化数据传输时延。该方法考虑了云数据中心间的带宽、私有云数据中心个数和容量等因素对传输时延的影响;通过引入遗传算法的交叉算子和变异算子,避免了粒子群优化算法的过早收敛,提高了种群进化的多样性,有效地压缩了数据传输时延。通过实验证明了所提算法的有效性。

关键词: 混合云, 科学工作流, 数据布局, 时延优化

Abstract: It is a major challenge to combine the advantages of both public and private clouds to rationalize the data placement of scientific workflow with private data,and optimize the transmission time of large-scale data across different data centers.By considering the characteristics of data placement in hybrid cloud and combining the dependencies among scientific workflow,an adaptive Discrete Particle Swarm Optimization Algorithm based on Genetic Algorithm operators (GA-DPSO) was proposed,which considered the influence on the transmission time such as the bandwidth between data centers,the number and the capacity of private cloud data centers.Through introducing the crossover operator and the mutation operator of the genetic algorithm,the premature convergence problem of the particle swarm optimization was avoided,which enhanced the diversity of population evolution and effectively compressed data transmission time.The experimental results showed that the data placement strategy based on GA-DPSO could effectively reduce the data transmission time of scientific workflow in hybrid cloud.

Key words: hybrid cloud, scientific workflow, data placement, time-driven optimization

中图分类号: