计算机科学 ›› 2024, Vol. 51 ›› Issue (12): 20-29.doi: 10.11896/jsjkx.240300064
粘英璞1, 易波1, 李沛辰1, 王兴伟1, 黄敏2
NIAN Yingpu1, YI Bo1, LI Peichen1, WANG Xingwei1, HUANG Min2
摘要： 当前,知识定义网络赋能AI技术发展,算力网络提供AI所需算力资源,二者逐渐趋于融合,形成了知识定义算力网络(Knowledge Defined Computing Networking,KDCN)。KDCN赋能发展了诸多新型网络应用,如元宇宙、AR/VR、东数西算等,这些新型应用对算力资源和网络资源有极大的需求,被称为重击流(Heavy Hitter,HH)。HH流的存在严重加剧了KDCN网络的拥塞情况。针对这一挑战,提出了一种智能流量调度机制,旨在通过深度Q神经网络来解决KDCN中的拥塞问题。相较于离线训练过程,通过流量数据检测与采集、在模型训练和拥塞流调决策之间建立实时闭环,来实现深度Q神经网络模型的在线训练。基于该闭环控制,智能流调模型通过不断学习可以实现持续演化,并用于提供实时决策。实验结果表明,该算法在资源利用率、吞吐量、平均丢包率等方面优于现有方法。
中图分类号:
| [1]Cisco.White Paper.Cisco Annual Internet Report 2018－2023[R/OL]Mar 2020.https://www.cisco.com/c/en/us/solutions/collateral/executive-perspectives/annual-internet-report/white-paper-c11-741490.html. [2]MORANDIN G M.Transmission Control Protocol－TCP[OL].https://www.zhangqiaokeyan.com/patent-detail/06130437989464.html. [3]LATIF Z,SHARIF K,LI F,et al.A Comprehensive Survey of Interface Protocols for Software Defined Networks[J].Journal of Network and Computer Applications,2020,156:102563. [4]MOIN S,KARIM A,SAFDAR K,et al.GREEN SDN－An Enhanced Paradigm of SDN:Review,Tax-onomy,and Future Directions[J].Concurrency and Computation:Practice and Experience,2020,32(21):e5086. [5]ALBERT M,RODRIGUEZ-NATAL A,CARNER J,et al.Knowledge-Defined Networking[J].CM SIGCOM Computer Communication Review,2017,47(3):2-10. [6]CLARK D.A knowledge plane for the internet[C]//Procee-dings of the 2003 Conference on Applications,Technologies,Architectures,and Protocols for Computer Communications.2003:3-10. [7]DALAL S,DAHIYA N,SETH B,et al.Adaptive Traffic Rou-ting Practice for Load Balance and Congestion Control in AdHoc Network in Cloud-MANET[J/OL].https://www.researchgate.net/publication/353498753_Adaptive_Traffic_Routing_Practice_for_Load_Balance_and_Congestion_Control_in_AdHoc_Network_in_Cloud-MANET. [8]ZHANG L,CUI Y,WANG M,et al.DeepCC:Bridging the Gap Between Congestion Control and Applications via Multiobjective Optimization[J].IEEE/ACM Transactions on Networking,2022,30(5):2274-2288. [9]WANG Y,WANG L,DONG X.An Intelligent TCP Congestion Control Method Basedon Deep Q Network[J].Future Internet,2021,13(10):261. [10]HM A,DU X A,YD B,et al.An intelligent scheme for conges-tion control:Whenactive queue management meets deep reinforcement learning[J].Computer Networks,2021,200:108515. [11]ALIZADEH M,EDSALL T,DHARMAPURIKAR S,et al.CONGA:distributed congestion-aware load balancing for datacenters[J].ACM SIGCOMM Computer Communication Review.2041,44(4):503-514. [12]MOHAMMAD A,SIVASANKAR R,BARATH R,et al.He-dera:dynamic flow scheduling for data center networks[C]//Proceedings of the 7th USENIX Conference on Networked Systems Design and Implementation(NSDI’10).USENIX Association.2010. [13]ZAHER M,ALAWADI A,MOLNAR S.Class-based FlowScheduling Framework in SDN-based Data Center Networks[C]//International Conference on Computing,Electronics & Communications Engineering(iCCECE).2020. [14]LIU Z,LI K,WU L,et al.CATS:Cost Aware Task Scheduling in Multi-Tier Computing Networks[J].Journal of Computer Research and Development,2020,57(9):1810-1822. [15]AMPRATWUM I.An Intelligent Traffic ClassificationBasedOptimized Routing in SDN-IoT:A Machine Learning Approach[D].Ottawa:University of Ottawa,2020. [16]MOUFAKIR T,ZHANI M F,GHERBI A,et al,Collaborative Multi-domain Routing in S DN Environments[J].Journal of Network and Systems Management,2022,30(1):1-23. [17]VAN ROSSEMS,TAVERNIER W,SONKOLY B,et al.Deploying elastic routing capability in an SDN/NFV-enabled environment[C]//IEEE Conference on Network Function Virtualization and Software Defined Network(NFV-SDN).2015:22-24. [18]SHI X,LI Q,WANG D,et al.Mobile Computing Force Network(MCFN):Computing and Network Convergence Supporting Integrated Communication Service[C]//International Conference on Service Science(ICSS).2022:131-136. [19]RUELAS A M R,ROTHENBERG C E.A Load BalancingMethod based on Artificial Neural Networks for Knowledge-defined Data Center Networking[C]//IFIP LANC.2018. [20]PHAM Q T A,HADJADJ-AOUL Y,OUTTAGARTS A.Deep Reinforcement Learning Based QoS-Aware Routing in Know-ledge-Defined Networking[M]//Social Informatics and Telecommunications Engineering.2019:14-26. [21]DUQUE-TORRES A,AMEZQUITA-SUAREZ F,RENDONO,et al.An Approach Based on Knowledge-Defined Networking for Identifying Heavy-Hitter Flows in Data Center Networks[J].Applied Sciences,2019,9(22):1-19. [22]SIMONYAN K,ZISSERMAN A.Very Deep Convolutional Networks for Large-Scale Image Recognition[J].arXiv:1409.1556,2014. [23]JIAWEI H,MICHELINE K.Data Mining:Concepts and Techniques[J].Data Mining C oncepts Models Methods & Algorithms Second Edition,2006,5(4):1-18. [24]JANE M,GREG G.Editorial for Journal of Field RoboticsSpecial Issue on Machine Learning Based Robotics in Unstructured Environments[J].Journal of Field Robotics,2006,23(11/12):943-944. [25]DUAN Y,NI H,ZHU X,et al.A Single-Rate Multicast Congestion Control(SRMCC) Mechanism in Information-Centric Networking[J].Future Internet,2022,14(2):38. [26]LEE C,JUNG J,CHUNG J M.DEFT:Multipath TCP for High Speed Low Latency Communications in 5G Networks[J].IEEE Transactions on Mobile Computing,2020,20(12):3311-3321. |
|
||
首页