Author Login Editor-in-Chief Peer Review Editor Work Office Work

Computer Engineering ›› 2021, Vol. 47 ›› Issue (8): 124-130,139. doi: 10.19678/j.issn.1000-3428.0058521

• Advanced Computing and Data Processing • Previous Articles     Next Articles

Data Replication Strategy of Cloud Storage Based on Coordinator and Genetic Algorithm

WEI Xiuran1, WANG Feng2   

  1. 1. College of Information and Management Science, Henan Agricultural University, Zhengzhou 450046, China;
    2. College of Information Engineering, North China University of Water Resources and Electric Power, Zhengzhou 450045, China
  • Received:2020-06-02 Revised:2020-07-22 Published:2020-08-04

基于协调器与遗传算法的云存储数据复制策略

魏秀然1, 王峰2   

  1. 1. 河南农业大学 信息与管理科学学院, 郑州 450046;
    2. 华北水利水电大学 信息工程学院, 郑州 450045
  • 作者简介:魏秀然(1975-),女,实验师、硕士,主研方向为云计算、智能算法;王峰,副教授、硕士。
  • 基金资助:
    河南省重点科技攻关项目(152102210112);河南省教育厅科学技术研究重点项目(13A520713);2017年国家大学生创新创业训练项目(201710466046)。

Abstract: For the optimization of cloud-based data storage, this paper proposes a new data replication strategy combining with a coordinator and the Genetic Algorithm(GA). The coordinator is built based on Hadoop Distributed File System(HDFS) architecture for replication management. The GA is used along with the query receiving algorithm to receive the queries and send them to the appropriate nodes to meet users' expected Quality of Service(QoS) requirements. At the same time, the physical location of the data block in a query is considered to obtain better replication parameters. The simulation results show that, compared with the existing typical strategies for data center selection and dynamic data replication strategy, as well as the gradual data deletion and addition, this data replication strategy not only optimizes the system load allocation, but also has higher availability and less latency.

Key words: cloud storage, data replication, coordinator, Hadoop Distributed File System(HDFS), query algorithm, Genetic Algorithm(GA), Quality of Service(QoS)

摘要: 针对云存储数据过程,结合协调器与遗传算法提出一种新的数据复制策略。在Hadoop分布式文件系统体系结构基础上构建一个用于复制管理的协调器,采用接收查询算法和遗传算法接收查询,并将其发送给合适的节点以满足用户期望的服务质量功能需求,同时考虑一个查询中数据块的物理位置以获得更好的复制参数。仿真结果表明,与目前典型的数据中心选择和动态数据复制策略以及逐步删除和添加数据副本策略相比,该数据复制策略不仅优化了系统的负荷分配,而且具有更高的可用性和更小的延迟。

关键词: 云存储, 数据复制, 协调器, Hadoop分布式文件系统, 查询算法, 遗传算法, 服务质量

CLC Number: