计算机科学

计算机科学 ›› 2024, Vol. 51 ›› Issue (2): 300-310.doi: 10.11896/jsjkx.230600128

• 计算机网络 • 上一篇    下一篇

面向缓存的动态协作任务迁移技术研究

赵晓焱1,2, 赵斌1, 张俊娜1,2, 袁培燕1   

  1. 1 河南师范大学计算机信息与工程学院 河南 新乡453007
    2 智慧商务与物联网技术河南省工程实验室 河南 新乡453007
  • 收稿日期:2023-06-15 修回日期:2023-09-26 出版日期:2024-02-15 发布日期:2024-02-22
  • 通讯作者: 张俊娜(jnzhang@htu.edu.cn)
  • 作者简介:(zhaoxiaoyan@htu.edu.cn)
  • 基金资助:
    国家自然科学基金(62072159,61902112);河南省科技攻关项目(222102210011,232102211061)

Study on Cache-oriented Dynamic Collaborative Task Migration Technology

ZHAO Xiaoyan1,2, ZHAO Bin1, ZHANG Junna1,2, YUAN Peiyan1   

  1. 1 College of Computer and Information Engineering,Henan Normal University,Xinxiang,Henan 453007,China
    2 Engineering Lab of Intelligence Business&Internet of Things,Xinxiang,Henan 453007,China
  • Received:2023-06-15 Revised:2023-09-26 Online:2024-02-15 Published:2024-02-22
  • About author:ZHAO Xiaoyan,born in 1981,Ph.D,associate professor,is a member of CCF(No.K8282M).Her main research interests include mobile edge computing,D2D communication and Internet of things.ZHANG Junna,born in 1979,Ph.D,associate professor,is a member of CCF(No.D2234M).Her main research interests include mobile edge computing and services computing.
  • Supported by:
    National Natural Science Foundation of China (62072159,61902112) and Science and Technology Research Project of Henan Province(222102210011,232102211061).

PDF (PC)

930

摘要： 边缘网络中不断出现的计算密集和延迟敏感型业务推动了任务迁移技术的快速发展。然而,任务迁移过程中存在应用场景复杂多变、问题建模难度高等技术瓶颈。尤其是考虑用户移动时,如何保证用户服务的稳定性和连续性,设计合理的任务迁移策略仍是一个值得深入探讨的问题。因此,提出了一种移动感知的服务预缓存模型和任务预迁移策略,将任务迁移问题转化为最优分簇与边缘服务预缓存的组合优化问题。首先,基于用户的移动轨迹对当前执行任务状态进行预测,引入动态协作簇和迁移预测半径的概念,提出了一种面向移动和负载两种任务场景的预迁移模型,解决了何时何地迁移的问题。然后,针对需要迁移的任务,基于最大容忍时延约束分析协作簇半径和簇内目标服务器数量的极限值,提出了以用户为中心的分布式多服务器间动态协作分簇算法(Distributed Dynamic Multi-server Cooperative Clustering Algorithm,DDMC)以及面向服务缓存的深度强化学习算法(Cache Based Double Deep Q Network,C-DDQN),解决了最优分簇和服务缓存问题。最后,利用服务缓存的因果关系,设计了一种低复杂度的交替最小化服务缓存位置更新算法,求解出了最佳迁移目标服务器集合,实现了任务迁移中的服务器协作及网络负载均衡。实验结果表明,提出的迁移选择算法具有良好的鲁棒性和系统性能,相比其他迁移算法所消耗的总成本降低了至少12.06%,所消耗的总时延降低了至少31.92%。

关键词: 移动边缘计算, 服务缓存, 动态协作簇, 任务迁移, 深度强化学习

Abstract: Task migration technology has been propelled by the continuous emergence of compute-intensive and delay-sensitive services in edge networks.However,the process of task migration is hindered by technical bottlenecks such as complex and time-varying application scenarios,as well as the high difficulty in problem modeling.Especially when considering user movement,designing a reasonable task migration strategy that ensures the stability and the continuity of user service remains a persistent challenge.Therefore,a mobile-aware service pre-caching model and task pre-migration strategy are proposed to transform the problem of task migration into an optimization problem that combines optimal clustering strategies with edge service pre-caching.First of all,the current state of the task is initially predicted based on the user′s movement trajectory.To solve the problem of when and where to migrate,a pre-migration model for two task scenarios,namely mobile and load,is proposed by introducing the concept of dynamic cooperation cluster and migration prediction radius.And then,according to the tasks that need to be migrated,the maximum tolerant delay constraint is utilized to derive the limit value of cooperative cluster radius and target server quantity in a cluster.Subsequently,a user-centric distributed dynamic multi-server cooperative clustering algorithm(DDMC) and a cache-based double deep Q network algorithm(C-DDQN) for service are proposed to solve the problem of optimal clustering and service ca-ching.Finally,a low-complexity alternate minimization service cache location update algorithm is designed using the causality of service caches to achieve the optimal set of migration target servers,which realize server collaboration and network load balancing in task migration.Experimental results demonstrate the robustness and the system performance of the proposed migration selection algorithm.Compared with other algorithms,the total cost consumed is reduced by at least 12.06%,the total latency consumed is reduced by at least 31.92%.

Key words: Mobile edge computing, Service cache, Dynamic collaborative cluster, Task migration, Deep reinforcement learning

中图分类号: 

  • TP393
[1]ZHANG J,HU X,NING Z,et al.Joint resource allocation for latency-sensitive services over mobile edge computing networks with caching [J].IEEE Internet of Things Journal,2018,6(3):4283-4294.
[2]MIAO Y,WU G,LI M,et al.Intelligent task prediction andcomputation offloading based on mobile-edge cloud computing [J].Future Generation Computer Systems,2020,102:925-931.
[3]ZHANG N,GUO S,DONG Y,et al.Joint task offloading and data caching in mobile edge computing networks [J].Computer Networks,2020,182:107446.
[4]PENG K,NIE J,KUMAR N,et al.Joint optimization of service chain caching and task offloading in mobile edge computing [J].Applied Soft Computing,2021,103:107142.
[5]LI C,ZHANG Y,GAO X,et al.Energy-latency tradeoffs for edge caching and dynamic service migration based on DQN in mobile edge computing [J].Journal of Parallel and Distributed Computing,2022,166:15-31.
[6]TANG F,LIU C,LI K,et al.Task migration optimization forguaranteeing delay deadline with mobility consideration in mobile edge computing [J].Journal of Systems Architecture,2020,112(8):101849.
[7]LI C,ZHU L,LI W,et al.Joint edge caching and dynamic service migration in SDN based mobile edge computing [J].Journal of Network and Computer Applications,2021,177:102966.
[8]OUYANG T,ZHOU Z,CHEN X.Follow me at the edge:Mo-bility-aware dynamic service placement for mobile edge computing [J].IEEE Journal on Selected Areas in Communications,2018,36(10):2333-2345.
[9]GE S,CHENG M,HE X,et al.A two-stage service migration algorithm in parked vehicle edge computing for internet of things [J].Sensors,2020,20(10):2786.
[10]YIN L,LI P,LUO J.Smart contract service migration mechanism based on container in edge computing [J].Journal of Pa-rallel and Distributed Computing,2021,152:157-166.
[11]WANG S,URGAONKAR R,ZAFER M,et al.Dynamic service migration in mobile edge computing based on Markov decision process [J].IEEE/ACM Transactions on Networking,2019,27(3):1272-1288.
[12]BI S,HUANG L,ZHANG Y J A.Joint optimization of service caching placement and computation offloading in mobile edge computing systems [J].IEEE Transactions on Wireless Communications,2020,19(7):4947-4963.
[13]XIE Q,WANG Q,YU N,et al.Dynamic service caching in mobile edge networks[C]//2018 IEEE 15th International Confe-rence on Mobile Ad Hoc and Sensor Systems(MASS).IEEE,2018:73-79.
[14]ZHAO T,HOU I H,WANG S,et al.Red/led:An asymptotically optimal and scalable online algorithm for service caching at the edge [J].IEEE Journal on Selected Areas in Communications,2018,36(8):1857-1870.
[15]CHEN L,XU J,REN S,et al.Spatio－temporal edge service placement:A bandit learning approach [J].IEEE Transactions on Wireless Communications,2018,17(12):8388-8401.
[16]ZHANG T,FAN H,LOO J,et al.User preference aware caching deployment for device-to-device caching networks [J].IEEE Systems Journal,2017,13(1):226-237.
[17]PENG T,WANG H,LIANG C,et al.Value-aware cache re-placement in edge networks for Internet of Things [J].Transactions on Emerging Telecommunications Technologies,2021,32(9):e4261.
[18]ZHANG W,WU D,YANG W,et al.Caching on the move:A user interest-driven caching strategy for D2D content sharing [J].IEEE Transactions on Vehicular Technology,2019,68(3):2958-2971.
[19]NAIR V,HINTON G E.Rectified linear units improve restric-ted boltzmann machines[C]//Proceedings of the 27th international conference on machine learning(ICML-10).2010:807-814.
[20]TALEB T,KSENTINI A,FRANGOUDIS P A.Follow-mecloud:When cloud services follow mobile users[J].IEEE Transactions on Cloud Computing,2016,7(2):369-382.
[21]MICHAEL M M,SCOTT M L.Nonblocking algorithms andpreemption-safe locking on multiprogrammed shared memory multiprocessors[J].Journal of parallel and distributed computing,1998,51(1):1-26.
[22]MEIZHEN W,YANLEI S,YUE T.The design and implementation of LRU-based web cache[C]//2013 8th International Conference on Communications and Networking in China(CHINACOM).IEEE,2013:400-404.
[23]SOKOLINSKY L B.LFU-K:An effective buffer managementreplacement algorithm[C]//Database Systems for Advanced Applications:9th International Conference(DASFAA 2004).Berlin Heidelberg:Springer,2004:670-681.
[24]CAO P,IRANI S.Cost-aware www proxy caching algorithms[C]//Usenix Symposium on Internet Technologies and Systems.1997:193-206.
[25]MNIH V,KAVUKCUOGLU K,SILVER D,et al.Human-level control through deep reinforcement learning [J].Nature,2015,518(7540):529-533.
[26]LI C,SONG M,ZHANG M,et al.Effective replica management for improving reliability and availability in edge-cloud computing environment [J].Journal of Parallel and Distributed Computing,2020,143:107-128.
[27]LI C,CAI Q,LOU Y.Optimal data placement strategy considering capacity limitation and load balancing in geographically distributed cloud [J].Future Generation Computer Systems,2022,127:142-159.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发