[1]梁 杨,丁长松,胡志刚.基于“推荐-学习”的两阶段数据布局策略[J].南京师大学报(自然科学版),2023,46(04):80-90.[doi:10.3969/j.issn.1001-4616.2023.04.012]
　Liang Yang,Ding Changsong,Hu Zhigang.A Two-Stage Data Placement Strategy Based on Recommendation and Learning[J].Journal of Nanjing Normal University(Natural Science Edition),2023,46(04):80-90.[doi:10.3969/j.issn.1001-4616.2023.04.012]
点击复制

基于“推荐-学习”的两阶段数据布局策略()
分享到：

《南京师大学报（自然科学版）》[ISSN:1001-4616/CN:32-1239/N]

卷:
第46卷
期数:
2023年04期
页码:
80-90
栏目:
计算机科学与技术
出版日期:
2023-12-15

文章信息/Info

Title:
A Two-Stage Data Placement Strategy Based on Recommendation and Learning
文章编号:
1001-4616(2023)04-0080-11
作者:
梁 杨1; 2; 丁长松1; 2; 胡志刚3
(1.湖南中医药大学信息科学与工程学院,湖南 长沙 410208)
(2.湖南省中医药大数据分析实验室,湖南 长沙 410208)
(3.中南大学计算机学院,湖南 长沙 410083)
Author(s):
Liang Yang1; 2; Ding Changsong1; 2; Hu Zhigang3
(1.School of Informatics,Hunan University of Chinese Medicine,Changsha 410208,China)
(2.Big Data Analysis Laboratory of TCM in Hunan Province,Changsha 410208,China)
(3.School of Computer Science and Engineering,Central South University,Changsha 41
关键词:
数据布局; 服务质量; 数据密集型; 副本推荐; 规则学习
Keywords:
data placement;  quality of service;  data-intensive;  replica recommendation;  rule learning
分类号:
TP391
DOI:
10.3969/j.issn.1001-4616.2023.04.012
文献标志码:
A
摘要:
针对数据布局不合理导致云边协同集群服务质量下降和运营开销增加等问题,提出一种基于“推荐-学习”的两阶段数据副本管理机制.鉴于数据密集型移动应用的特点,综合考虑了云边环境下的数据访问延迟和放置代价之间的最优权衡.从理论上构建了副本放置的多目标数学模型,将决策问题描述为具有延迟和成本约束的双目标优化问题; 在“推荐”阶段,设计了一个基于移动预测和反馈优化的副本推荐引擎,减少了副本创建的盲目性; 在“学习”阶段,构建了一个基于异步优势行动者-评论家算法(A3C)的强化学习副本放置规则学习模型,改进了副本服务的全局性能指标.实验结果表明,基于“推荐-学习”的两阶段数据布局策略能够有效地减少等待延迟和节约成本开销,为现代云边协同系统的数据管理服务提供行之有效的解决方案,具有重要的理论意义和应用价值.
Abstract:
Aiming at the problem that the unreasonable data layout leads to the decrease of service quality and the increase of operation cost in cloud-edge collaborative cluster,a two-stage data replica management mechanism based on ‘recommendation and learning'method is proposed. In view of the characteristics of data-intensive mobile applications,the optimal trade-off between data access latency and placement cost in the cloud environment is considered. The multi-objective mathematical model of replica placement is established in theory,and the decision problem is described as a bi-objective optimization problem with delay and cost constraints. In the recommendation stage,a replica recommendation engine based on motion prediction and feedback optimization is proposed to reduce the blindness of replica creation. In the learning stage,a reinforcement learning replica placement rule learning model based on Asynchronous Advantage Actor-Critic algorithm(A3C)is constructed to improve the global performance index of replica service. The experimental results show that the two-stage data placement strategy based on ‘recommendation and learning' method can reduce latency time and save cost overhead effectively and efficiently. Meanwhile,our proposed method provides an effective solution for challenges of data managemen in the real-life cloud-edge collaboration systems,which has important theoretical significance and application value.

参考文献/References:

[1]WANG T,ZHAO D,CAI S B,et al. Bidirectional prediction-based underwater data collection protocol for end-edge-cloud orchestrated system[J]. IEEE transactions on industrial informatics,2019,16(7):4791-4799.
[2]施巍松,张星洲,王一帆,等. 边缘计算:现状与展望[J]. 计算机研究与发展,2019,56(1):69-89.
[3]LI A T,CHEN Y,YAN Z,et al. A survey on integrity auditing for data storage in the cloud:from single copy to multiple replicas[J]. IEEE transactions on big data,2022,8(5):1428-1442.
[4]CHIANG M L,HSIEH H C,CHANG T Y,et al. An adaptive replica configuration mechanism based on predictive file popularity and queue balance in mobile edge computing environment[J]. Soft computing,2023,27(1):107-129.
[5]AGIWAL M,ROY A,SAXENA N. Next generation 5G wireless networks:a comprehensive survey[J]. IEEE communications surveys & tutorials,2016,18(3):1617-1655.
[6]ASIM M,WANG Y,WANG K,et al. A review on computational intelligence techniques in cloud and edge computing[J]. IEEE transactions on emerging topics in computational intelligence,2020,4(6):742-763.
[7]徐光伟,史春红,冯向阳,等. 基于多级网络编码的多副本云数据存储[J]. 计算机研究与发展,2021,58(2):293-304.
[8]HE S W,REN J,WANG J H,et al. Cloud-edge coordinated processing:low-latency multicasting transmission[J]. IEEE journal on selected areas in communications,2019,37(5):1144-1158.
[9]AFONSO N,BRAVO M,RODRIGUES L. Combining high throughput and low migration latency for consistent data storage on the edge[C]//Proceedings of the 29th Int Conf on Computer Communications and Networks. Piscataway,NJ:IEEE,2020.
[10]LIU J W,SHEN H Y,NARMAN H S. Popularity-aware multi-failure resilient and cost-effective replication for high data durability in cloud storage[J]. IEEE transactions on parallel and distributed systems,2019,30(10):2355-2369.
[11]ARAL A,OVATMAN T. A decentralized replica placement algorithm for edge computing[J]. IEEE transactions on network and service management,2018,15(2):516-529.
[12]LI C L,WANG Y P,TANG H L,et al. Dynamic multi-objective optimized replica placement and migration strategies for SaaS applications in edge cloud[J]. Future generation computer systems,2019,100:921-937.
[13]JIN J H,LI Y H,LUO J Z. Cooperative storage by exploiting graph-based data placement algorithm for edge computing environment[J]. Concurrency and computation:practice and experience,2018,30(20):1049-1064.
[14]VALES R,MOURA J,MARINHEIRO R. Energy-aware and adaptive fog storage mechanism with data replication ruled by spatio-temporal content popularity[J]. Journal of network and computer applications,2019,135:84-96.
[15]CHANG W C,WANG P C. Adaptive replication for mobile edge computing[J]. IEEE journal on selected areas in communica-tions,2018,36(11):2422-2432.
[16]TERANISHI Y,KIMATA T,YAMANAKA H,et al. Dynamic data flow processing in edge computing environments[C]//Proceedings of the 41st International Computer Software and Applications Conference. Piscataway,NJ:IEEE,2017:935-944.
[17]XING J R,DAI H J,YU Z L. A distributed multi-level model with dynamic replacement for the storage of smart edge computing[J]. Journal of systems architecture,2018,83:1-11.
[18]ARAL A,OVATMAN T. A decentralized replica placement algorithm for edge computing[J]. IEEE transactions on network and service management,2018,15(2):516-529.
[19]WANG J D,ZHAO L,LIU J J,et al. Smart resource allocation for mobile edge computing:a deep reinforcement learning approach[J]. IEEE transactions on emerging topics in computing,2021,9(3):1529-1541.
[20]DAI P L,HANG Z H,LIU K,et al. Multi-armed bandit learning for computation-intensive services in mec-empowered vehicular networks[J]. IEEE transactions on vehicular technology,2020,69(7):7821-7834.
[21]TEREFE M B,LEE H,HEO N,et al. Energy-efficient multisite offloading policy using Markov decision process for mobile cloud computing[J]. Pervasive and mobile computing,2016,27:75-89.
[22]WU H,LUO Y L,LI C L. Optimization of heat-based cache replacement in edge computing system[J]. Journal of supercomputing,2021,77(3):2268-2301.
[23]SONMEZ C,OZGOVDE A,ERSOY C. EdgeCloudSim:an environment for performance evaluation of edge computing systems[C]//Proceedings of the 2nd Int Conf on Fog and Mobile Edge Computing. Piscataway,NJ:IEEE,2017:39-44.
[24]MEDINA A,LAKHINA A,MATTA I,et al. BRITE:An approach to universal topology generation[C]//Proceedings of the 9th Int Symp on Modeling,Analysis and Simulation of Computer and Telecommunication Systems. Piscataway,NJ:IEEE,2001:346-353.
[25]TANG Y Y,WANG H,GUO K H,et al. A new replica placement mechanism for mobile media streaming in edge computing[J]. Concurrency and computation:practice and experience,2019,33(7):e5361.
[26]MANSOURI N,JAVIDI M. A hybrid data replication strategy with fuzzy-based deletion for heterogeneous cloud data centers[J]. Journal of supercomputing,2018,74(10):5349-5372.

备注/Memo

备注/Memo:
收稿日期:2023-05-30.
基金项目:国家自然科学基金项目(62172442)、湖南省教育厅优秀青年项目(22B0400)、湖南中医药大学校级科研基金(2021XJJJ021)、湖南省自然科学基金项目(2023JJ60124)、长沙市科技局项目(kq2202265)、湖南中医药大学学科建设“揭榜挂帅”项目(22JBZ049).
通讯作者:胡志刚,博士,教授,博士生导师,研究方向:高性能计算、云计算、边缘计算和能效资源管理. E-mail:zghu@csu.edu.cn
更新日期/Last Update: 2023-12-15