Extendable NFV-Integrated Control Method Using Reinforcement Learning

被引:10
作者
Suzuki, Akito [1 ]
Kawahara, Ryoichi [2 ]
Kobayashi, Masahiro [1 ]
Harada, Shigeaki [1 ]
Takahashi, Yousuke [1 ]
Ishibashi, Keisuke [3 ]
机构
[1] NTT Corp, NTT Network Technol Labs, Musashino, Tokyo 1808585, Japan
[2] Toyo Univ, Fac Informat Networking Innovat & Design, Tokyo 1150053, Japan
[3] Int Christian Univ, Div Arts & Sci, Coll Liberal Arts, Mitaka, Tokyo 1810015, Japan
关键词
NFV; network control; reinforcement learning; CHALLENGES; MANAGEMENT;
D O I
10.1587/transcom.2019EBP3114
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Network functions virtualization (NFV) enables telecommunications service providers to realize various network services by flexibly combining multiple virtual network functions (VNFs). To provide such services, an NFV control method should optimally allocate such VNFs into physical networks and servers by taking account of the combination(s) of objective functions and constraints for each metric defined for each VNF type, e.g., VNF placements and routes between the VNFs. The NFV control method should also be extendable for adding new metrics or changing the combination of metrics. One approach for NFV control to optimize allocations is to construct an algorithm that simultaneously solves the combined optimization problem. However, this approach is not extendable because the problem needs to be reformulated every time a new metric is added or a combination of metrics is changed. Another approach involves using an extendable network-control architecture that coordinates multiple control algorithms specified for individual metrics. However, to the best of our knowledge, no method has been developed that can optimize allocations through this kind of coordination. In this paper, we propose an extendable NFV-integrated control method by coordinating multiple control algorithms. We also propose an efficient coordination algorithm based on reinforcement learning. Finally, we evaluate the effectiveness of the proposed method through simulations.
引用
收藏
页码:826 / 841
页数:16
相关论文
共 35 条
  • [1] Amaldi E., 2016, Electron. Notes Discr. Math., V52, P213, DOI DOI 10.1016/J.ENDM.2016.03.028
  • [2] Hybrid scheduling for the parallel solution of linear systems
    Amestoy, PR
    Guermouche, A
    L'Excellent, JY
    Pralet, S
    [J]. PARALLEL COMPUTING, 2006, 32 (02) : 136 - 156
  • [3] A fully asynchronous multifrontal solver using distributed dynamic scheduling
    Amestoy, PR
    Duff, IS
    L'Excellent, JY
    Koster, J
    [J]. SIAM JOURNAL ON MATRIX ANALYSIS AND APPLICATIONS, 2001, 23 (01) : 15 - 41
  • [4] [Anonymous], 2016, PROC INT C MACH LEAR
  • [5] [Anonymous], 2016, COMP COMM IEEE INFOC
  • [6] [Anonymous], 1998, INTRO REINFORCEMENT
  • [7] Coordinated Allocation of Service Function Chains
    Beck, Michael Till
    Felipe Botero, Juan
    [J]. 2015 IEEE GLOBAL COMMUNICATIONS CONFERENCE (GLOBECOM), 2015,
  • [8] Bernal DavidE., 2018, COMPUTER AIDED CHEM, V44, P895, DOI DOI 10.1016/B978-0-444-64241-7.50144-0
  • [9] JASPER: Joint Optimization of Scaling, Placement, and Routing of Virtual Network Services
    Draexler, Sevil
    Karl, Holger
    Mann, Zoltan Adam
    [J]. IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT, 2018, 15 (03): : 946 - 960
  • [10] Network Function Virtualization: Challenges and Opportunities for Innovations
    Han, Bo
    Gopalakrishnan, Vijay
    Ji, Lusheng
    Lee, Seungjoon
    [J]. IEEE COMMUNICATIONS MAGAZINE, 2015, 53 (02) : 90 - 97