Efficient Network Slicing with Dynamic Resource Allocation

doi:10.12142/ZTECOM.202101003

ZTE Communications ›› 2021, Vol. 19 ›› Issue (1): 11-19.DOI: 10.12142/ZTECOM.202101003

• Special Topic • Previous Articles Next Articles

Efficient Network Slicing with Dynamic Resource Allocation

JI Hong¹(), ZHANG Tianxiang², ZHANG Kai¹, WANG Wanyuan¹, WU Weiwei¹

^1.School of Computer Science and Engineering, Southeast University, Nanjing 211189, China
^2.ZTE corporation, Shenzhen 518057, China

Received:2020-12-09 Online:2021-03-25 Published:2021-04-09
About author:JI Hong (hong_ji@seu.edu.cn) received the B.S. degree from School of Economics and Management from Xidian University, China in 2018. He is currently a master student at School of Cyber Science and Engineering, Southeast University, China. His research interests lie in network optimization and intelligent decision making.|ZHANG Tianxiang received the B.S. degree from Nanjing University of Aeronautics and Astronautics, China. He is currently an engineer with ZTE Corporation. His research interests include traffic scheduling and graph neural networks.|ZHANG Kai is currently a master student at School of Cyber Science and Engineering, Southeast University, China. His research interests include graph neural networks and resource allocation and reinforcement learning.|WANG Wanyuan is an assistant professor with the School of Computer Science and Engineering, Southeast University, China. He received his Ph. D. degree in computer science from Southeast University in 2016. He has published several articles in refereed journals and conference proceedings, such as the IEEE Transactions on Mobile Computing, IEEE Journal on Selected Areas in Communications, IEEE Transactions on Cybernetics, AAAI, and AAMAS. He won the best student paper award from ICTAI14. His main research interests include artificial intelligence, multiagent systems, and game theory.|WU Weiwei is a professor in the School of Computer Science and Engineering, Southeast University, China. He received his B.Sc. degree fron South China University of Technology，China and the Ph.D. degree from City University of Hong Kong (CityU)，China and University of Science and Technology of China (USTC) in 2011, and went to Nanyang Technological University, Singapore for post-doctorial research in 2012. He has published over 50 peer-reviewed papers in international conferences/journals, and serves as TPCs and reviewers for several top international journals and conferences. His research interests include optimizations and algorithm analysis, wireless communications, crowdsourcing, cloud computing, reinforcement learning, game theory and network economics.

Abstract

Abstract:

With the rapid development of wireless network technologies and the growing demand for a high quality of service (QoS), the effective management of network resources has attracted a lot of attention. For example, in a practical scenario, when a network shock occurs, a batch of affected flows needs to be rerouted to respond to the network shock to bring the entire network deployment back to the optimal state, and in the process of rerouting a batch of flows, the entire response time needs to be as short as possible. Specifically, we reduce the time consumed for routing by slicing, but the routing success rate after slicing is reduced compared with the unsliced case. In this context, we propose a two-stage dynamic network resource allocation framework that first makes decisions on the slices to which flows are assigned, and coordinates resources among slices to ensure a comparable routing success rate as in the unsliced case, while taking advantage of the time efficiency gains from slicing.

Key words: network slicing, dynamic resource allocation, reinforcement learning

JI Hong, ZHANG Tianxiang, ZHANG Kai, WANG Wanyuan, WU Weiwei. Efficient Network Slicing with Dynamic Resource Allocation[J]. ZTE Communications, 2021, 19(1): 11-19.

Figures/Tables 13

Table 1 Notation overview

Notation	Description
$V = {v 1, . . . ?, v n}$	the set of nodes
$E = {v s, ? v d i \| i = 1,2, . . . ?, m}$	the set of edges
$e i$ or $e s d$	a certain link
$A = {a 1, . . . ?, a m}$	available bandwidth per link
$G = {V, ? E, ? A}$	the set of a network
$F = {v s, ? v d j \| j = 1, . . . ?, h}$	the set of flows
$f j$ or $f s d$	a certain flow
$d j$	demand of a flow
$P j$	a path the flow $j$ takes
$S = {s 1, . . . ?, s k, . . . ?, s K}$	set of slices
$f j k ∈ F k$	set of flows to be deployed in slice $k$

Table 1 Notation overview

Notation	Description
$V = {v 1, . . . ?, v n}$	the set of nodes
$E = {v s, ? v d i \| i = 1,2, . . . ?, m}$	the set of edges
$e i$ or $e s d$	a certain link
$A = {a 1, . . . ?, a m}$	available bandwidth per link
$G = {V, ? E, ? A}$	the set of a network
$F = {v s, ? v d j \| j = 1, . . . ?, h}$	the set of flows
$f j$ or $f s d$	a certain flow
$d j$	demand of a flow
$P j$	a path the flow $j$ takes
$S = {s 1, . . . ?, s k, . . . ?, s K}$	set of slices
$f j k ∈ F k$	set of flows to be deployed in slice $k$

Figure 1 A network with 3 nodes and 3 edges

Figure 2 Network slices in our slicing model

Figure 3 Interaction between policy and environment

Figure 4 Reinforcement learning (RL) training architecture

Figure 5 Dynamic resource adjustment

Figure 6 An example of network topology of the ring network

Figure 7 Loss during training is based on a 2080 Ti graphics processing unit (GPU) and 120 sampling processes (2 Xeon CPUs)

Figure 8 Total reward of different methods

Figure 9 Success Rate of different methods

Figure 10 Relative failure number of dynamic resource adaptation vs. static slicing

Figure 11 Comparison of calculation time of different methods

Figure 12 Comparison of the effectiveness of combined methods

References 22

1	JEFFREY G A, BUZZI S, CHOI W, et al. What will 5G be? [J]. IEEE Journal on selected areas in communications, 2014, 32(6): 1065. DOI: 10.1109/JSAC.2014.2328098 DOI
2	EINSIEDLER H J, GAVRAS A, SELLSTEDT P, et al. System design for 5G converged networks [C]//2015 European Conference on Networks and Communications. Paris, France: IEEE, 2015: 391–396. DOI:10.1109/EuCNC.2015.7194105 DOI
3	HAWILO H, SHAMI A, MIRAHMADI M, et al. NFV: state of the art, challenges, and implementation in next generation mobile networks [J]. IEEE network, 2014, 28(6): 18–26. DOI:10.1109/MNET.2014.6963800 DOI
4	KARAKUS M, DURRESI A. Quality of service (QoS) in software defined networking (SDN): a survey [J]. Journal of network and computer applications, 2017, 80: 200–218. DOI: 10.1016/j.jnca.2016.12.019 DOI
5	WALLNER R, CANNISTRA R. An SDN approach: quality of service using big switch’s floodlight open⁃source controller [J]. Proceedings of the Asia⁃Pacific advanced network, 2013, 35: 14. DOI:10.7125/APAN.35.2 DOI
6	XU C, CHEN B, QIAN H. Quality of service guaranteed resource management dynamically in software defined network [J]. Journal of communications, 2015: 843–850. DOI:10.12720/jcm.10.11.843-850 DOI
7	XU C, GAMAGE S, LU H. vTurbo: accelerating virtual machine I/O processing using designated turbo⁃sliced core [C]//2013 USENIX Annual Technical Conference. San Jose, USA: USENIX, 2013:243–254
8	SCANO D, VALCARENGHI L, KONDEPU K, et al. Network slicing in SDN networks [C]//2020 22nd International Conference on Transparent Optical Networks (ICTON). Bari, Italy: IEEE, 2020: 1–4. DOI:10.1109/ICTON51198.2020.9203184 DOI
9	ZHU K, HOSSAIN E. Virtualization of 5G cellular networks as a hierarchical combinatorial auction [J]. IEEE transactions on mobile computing, 2016, 15(10): 2640–2654. DOI:10.1109/TMC.2015.2506578 DOI
10	KATSALIS K, NIKAEIN N, SCHILLER E, et al. Network slices toward 5G communications: Slicing the LTE network [J]. IEEE communications magazine, 2017, 55(8): 146–154. DOI:10.1109/MCOM.2017.1600936 DOI
11	BARI M F, BOUTABA R, ESTEVES R, et al. Data center network virtualization: A survey [J]. IEEE communications surveys & tutorials, 2013, 15(2): 909–928.DOI:10.1109/SURV.2012.090512.00043 DOI
12	ZHANG Q X, LIU F M, ZENG C B. Adaptive interference⁃aware VNF placement for service⁃customized 5G network slices [C]//IEEE INFOCOM 2019 IEEE Conference on Computer Communications. Paris, France: IEEE, 2019: 2449–2457. DOI:10.1109/INFOCOM.2019.8737660 DOI
13	CHEN J J, TSAI M H, ZHAO L Q, et al. Realizing dynamic network slice resource management based on SDN networks [C]//2019 International Conference on Intelligent Computing and its Emerging Applications (ICEA). Tainan, Taiwan, China: IEEE, 2019: 120–125. DOI:10.1109/ICEA.2019.8858288 DOI
14	PARSAEEFARD S, JUMBA V, DERAKHSHANI M, et al. Joint resource provisioning and admission control in wireless virtualized networks [C]//2015 IEEE Wireless Communications and Networking Conference. New Orleans, USA: IEEE, 2015: 2020–2025. DOI:10.1109/WCNC.2015.7127778 DOI
15	MONEMI M, RASTI M, HOSSAIN E. Low⁃complexity SINR feasibility checking and joint power and admission control in prioritized multitier cellular networks [J]. IEEE transactions on wireless communications, 2016, 15(3): 2421–2434. DOI:10.1109/TWC.2015.2504084 DOI
16	SUTTON R S, BARTO A G. Reinforcement learning: an introduction [M]. Cambridge, UK: MIT Press, 2018
17	SCHULMAN J, WOLSKI F, DHARIWAL P, et al. Proximal policy optimization algorithms [EB/OL]. (2017⁃08⁃28) [2021⁃12⁃09].
18	ESPEHOLT L, SOYER H, MUNOS R, et al. IMPALA: scalable distributed deep⁃RL with importance weighted actor⁃learner architectures [EB/OL]. (2018⁃06⁃28) [2021⁃12⁃09].
19	STOOKE A, ABBEEL P. Accelerated methods for deep reinforcement learning [EB/OL]. (2019⁃12⁃09) [2021⁃01⁃18].
20	MORITZ P, NISHIHARA R, WANG S, et al. Ray: A distributed framework for emerging AI applications [C]//13th USENIX Symposium on Operating Systems Design and Implementation. Carlsbad, USA: OSDI, 2018: 561–577
21	LIANG E, LIAW R, NISHIHARA R, et al. RLLIB: abstractions for distributed reinforcement learning [EB/OL]. (2018⁃07⁃29) [2021⁃12⁃09].
22	LIAW R, LIANG E, NISHIHARA R, et al. Tune: a research platform for distributed model selection and training [EB/OL]. (2018⁃07⁃13) [2021⁃12⁃09].

[1]	WEI Zhiqing, ZHANG Yongji, JI Danna, LI Chenfei. Sensing and Communication Integrated Fast Neighbor Discovery for UAV Networks [J]. ZTE Communications, 2024, 22(3): 69-82.
[2]	SHEN Jiahao, JIANG Ke, TAN Xiaoyang. Boundary Data Augmentation for Offline Reinforcement Learning [J]. ZTE Communications, 2023, 21(3): 29-36.
[3]	REN Min, XU Renyu, ZHU Ting. Double Deep Q-Network Decoder Based on EEG Brain-Computer Interface [J]. ZTE Communications, 2023, 21(3): 3-10.
[4]	FENG Bingyi, FENG Mingxiao, WANG Minrui, ZHOU Wengang, LI Houqiang. Multi-Agent Hierarchical Graph Attention Reinforcement Learning for Grid-Aware Energy Management [J]. ZTE Communications, 2023, 21(3): 11-21.
[5]	YU Junpeng, CHEN Yiyu. A Practical Reinforcement Learning Framework for Automatic Radar Detection [J]. ZTE Communications, 2023, 21(3): 22-28.
[6]	YOU Qian, XU Qian, YANG Xin, ZHANG Tao, CHEN Ming. RIS-Assisted UAV-D2D Communications Exploiting Deep Reinforcement Learning [J]. ZTE Communications, 2023, 21(2): 61-69.
[7]	JIA Haonan, HE Zhenqing, TAN Wanlong, RUI Hua, LIN Wei. Distributed Multi-Cell Multi-User MISO Downlink Beamforming via Deep Reinforcement Learning [J]. ZTE Communications, 2022, 20(4): 69-77.
[8]	ZHANG Jing, WEI Xiao, CHENG Junfeng, FENG Xu. Satellite E2E Network Slicing Based on 5G Technology [J]. ZTE Communications, 2020, 18(4): 26-33.
[9]	HAN Bin, Hans D. SCHOTTEN. Machine Learning for Network Slicing Resource Management:A Comprehensive Survey [J]. ZTE Communications, 2019, 17(4): 27-32.
[10]	Stephen ANOKYE, Mohammed SEID, SUN Guolin. A Survey on Machine Learning Based Proactive Caching [J]. ZTE Communications, 2019, 17(4): 46-55.
[11]	DONG Shaokang, CHEN Jiarui, LIU Yong, BAO Tianyi, GAO Yang. Reinforcement Learning from Algorithm Model to Industry Innovation: A Foundation Stone of Future Artificial Intelligence [J]. ZTE Communications, 2019, 17(3): 31-41.

Efficient Network Slicing with Dynamic Resource Allocation

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 13

References 22

Related Articles 11

Recommended Articles

Metrics