基于行为树的组网服务管控架构及调度方法

doi:10.12305/j.issn.1001-506X.2024.10.31

Abstract

Abstract:

In order to achieve end-to-end security, reliability, and real-time network business support, traditional internet requires the deployment of multiple types of networking service at network element nodes. However, most existing network devices only provide network access, routing and other service functions, lacking safe, reliable, real-time and other service design. Based on the open source network operating system software for open networking in the cloud (SONiC), a management and control architecture and scheduling method for networking service based on container technology and behavior tree is designed. It realizes the deployment of networking service for network service support, which is then deploy to programmable network devices for unified management and scheduling, and simulation testing is carried out. The networking service management and control architecture and scheduling method further enhance the capability of network equipment, and use the behavior tree to efficiently coordinate and schedule the status transformation in the management, control and scheduling process to ensure the delivery quality of networking service.

Key words: behavior tree, container technology, software for open networking in the cloud (SONiC), networking service scheduling

CLC Number:

TP273

Ruitao SONG, Xianming GAO, Jiaorui HUANG, Chungang YANG, Tao HUANG, Yuanyuan LI. Behavior tree based management and control architecture and scheduling method for networking service[J]. Systems Engineering and Electronics, 2024, 46(10): 3547-3556.

Figures/Tables 9

Fig.1

Table 1

Fig.2

Fig.3

Fig.4

Fig.5

Fig.6

Fig.7

Fig.8

References 15

16	POTDAR A M , NARAYAN D G , KENGOND S , et al. Perfor-mance evaluation of docker container and virtual machine[J]. Procedia Computer Science, 2020, 171, 1419- 1428. doi: 10.1016/j.procs.2020.04.152
17	AHMAD I , AIFAILAKAWI M G , AIMUTAWA A , et al. Container scheduling techniques: a survey and assessment[J]. Journal of King Saud University-Computer and Information Sciences, 2022, 34 (7): 3934- 3947. doi: 10.1016/j.jksuci.2021.03.002
18	OUYANG Y , LINZ J Q , FENGZ T , et al. Intent-driven CoX resource management for space-terrestrial networks[J]. IEEE Wireless Communications, 2023, doi: 10.1109/MWC.018.2200600
19	Mi X R , YANG C G , SONG Y B , et al. Matching game for intelligent resource management in integrated satellite-terrestrial networks[J]. IEEE Wireless Communications, 2022, 29 (6): 88- 94. doi: 10.1109/MWC.009.2100555
20	KAEWKASI C, CHUENMUNEEWONG K. Improvement of container scheduling for docker using ant colony optimization[C]// Proc. of the 9th International Conference on Knowledge and Smart Technology, 2017: 254-259.
21	MENOUER T . KCSS: Kubernetes container scheduling strategy[J]. The Journal of Supercomputing, 2021, 77 (5): 4267- 4293. doi: 10.1007/s11227-020-03427-3
22	李铭轩, 曹畅, 唐雄燕, 等. 基于可编程网络的UPF边缘调度机制研究[J]. 数据与计算发展前沿, 2022, 4 (2): 74- 86.
	LI M X , CAO C , TANG X Y , et al. Research on edge scheduling mechanism of UPF based on programmable network[J]. Frontiers of Data & Computing, 2022, 4 (2): 74- 86.
23	刘曦. 数据中心网络SONiC白盒技术的发展趋势[J]. 通信世界, 2019, (33): 42- 43.
	LIU X . The development trends of SONiC white box technology in data center networks[J]. Communication World, 2019, (33): 42- 43.
24	IOVⅡNO M , SCUKINS E , STYRUD J , et al. A survey of behavior trees in robotics and AI[J]. Robotics and Autonomous Systems, 2022, 154, 104096.
25	COLLEDANCHISE M. Behavior trees in robotics[D]. Stockholm: KTH Royal Institute of Technology, 2017.
26	PERGER A , GAMPER P , WITZMANN R . Behavior trees for smart grid control[J]. IFAC-PapersOnLine, 2022, 55 (9): 122- 127.
27	MARZINOTTO A, COLLEDANCHISE M, SMITH C, et al. Towards a unified behavior trees framework for robot control[C]// Proc. of the IEEE International Conference on Robotics and Automation, 2014: 5420-5427.
28	ROVIDA F, GROSSMANN B, KRUGER V. Extended beha-vior trees for quick definition of flexible robotic tasks[C]//Proc. of the IEEE/RSJ International Conference on Intelligent Robots and Systems, 2017: 6793-6800.
29	BIGGAR O , ZAMANI M , SHAMES I . An expressiveness hierarchy of behavior trees and related architectures[J]. IEEE Robotics and Automation Letters, 2021, 6 (3): 5397- 5404.
30	COLLEDANCHISE M, MARZINOTTO A, DIMAROGONAS D V, et al. The advantages of using behavior trees in mult-robot systems[C]//Proc. of the 47th International Symposium on Robotics, 2016.
31	STYRUD J, IOVINO M, NORRLOF M, et al. Combining planning and learning of behavior trees for robotic assembly[C]//Proc. of the International Conference on Robotics and Automation, 2022: 11511-11517.
32	WU R C, KORTIK S, SANTOS C H. Automated behavior tree error recovery framework for robotic systems[C]//Proc. of the IEEE International Conference on Robotics and Automation, 2021: 6898-6904.
33	CAI Z X, LI M L, HUANG W R, et al. BT expansion: a sound and complete algorithm for behavior planning of intelligent robots with behavior trees[C]//Proc. of the AAAI Conference on Artificial Intelligence, 2021.
1	POP M D, PANDEY J, RAMASAMY V. Future networks 2030: challenges in intelligent transportation systems[C]//Proc. of the 8th International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions), 2020: 898-902.
2	BONATI L , POLESE M , D'ORO S , et al. Open, programmable, and virtualized 5G networks: state-of-the-art and the road ahead[J]. Computer Networks, 2020, 182, 107516. doi: 10.1016/j.comnet.2020.107516
3	张成林, 宋玲玲. 面向未来网络的白盒交换机体系综述[J]. 信息技术与网络安全, 2022, 41 (3): 2- 8.
	ZHANG C L , SONG L L . A review on architecture of white box switches for future networks[J]. Information Technology & Network Security, 2022, 41 (3): 2- 8.
4	BOSSHART P , DALY D , GIBB G , et al. P4: programming protocol-independent packet processors[J]. ACM SIGCOMM Computer Communication Review, 2014, 44 (3): 87- 95. doi: 10.1145/2656877.2656890
5	Open Compute Project. Switch abstraction interface(SAI): a reference switch abstraction interface for OCP[EB/OL]. [2023-07-04]. https://www.opencompute.org/.
6	Open Compute Project. SONiC: software for open net-working in the cloud website[EB/OL]. [2023-07-04]. https://github.com/sonic-net/SONiC.
7	Open Compute Project. Open network linux: a reference network operating system for OCP[EB/OL]. [2023-07-04]. http://opennetlinux.org/.
8	ALAM I , SHARIF K , LI F , et al. A survey of network virtualization techniques for Internet of things using SDN and NFV[J]. ACM Computing Surveys, 2020, 53 (2): 1- 40.
9	SHEN X M , GAO J , WU W , et al. Holistic network virtualization and pervasive network intelligence for 6G[J]. IEEE Communications Surveys & Tutorials, 2021, 24 (1): 1- 30.
10	ZHANG L L, YANG C G, OUYANG Y, et al. ISFC: intent-driven service function chaining for satellite networks[C]//Proc. of the 27th Asia Pacific Conference on Communications, 2022: 544-549.
11	张露露, 杨春刚, 王栋, 等. 意图驱动的云网融合按需编排[J]. 电信科学, 2022, 38 (10): 107- 119.
	ZHANG L L , YANG C G , WANG D , et al. Intent-driven cloud-network convergence on-demand orchestration[J]. Telecommunications Science, 2022, 38 (10): 107- 119.
12	LONG X M, LIU B H, JIANG F, et al. FPGA virtualization deployment based on docker container technology[C]//Proc. of the 5th International Conference on Mechanical, Control and Computer Engineering, 2020: 473-476.
13	SIDDIQUI T, SIDDIQUI S A, KHAN N A. Comprehensive analysis of container technology[C]//Proc. of the 4th International Conference on Information Systems and Computer Networks, 2019: 218-223.
14	MUNERMAN V, MUNERMAN D. Realization of distributed data processing on the basis of container technology[C]//Proc. of the IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering, 2019: 1740-1744.
15	MI X R , YANG C G , SONG Y B , et al. A distributed matching game for exploring resource allocation in satellite networks[J]. Peer-to-Peer Networking and Applications, 2021, 14 (5): 3360- 3371. doi: 10.1007/s12083-021-01158-7

特性	有限状态机	分层状态机	决策树	行为树
模块化	状态单元	子状态机单元	决策单元	功能子树
层次化	数据驱动, 单向传输	分层定义, 子状态机之间分层处理	递归划分、层次逻辑决策/预测、结构可视化	时钟驱动, 分层执行逻辑节点, 双向传输
可读性	一般	一般	优	较好
可重用性	差	差	一般	较好
鲁棒性	(1) 状态复杂, 并行度低, 吞吐量低(2) 当逻辑状态较多时, 容易发生状态爆炸	(1) 子状态机并行运行, 容易发生冲突(2) 子状态机之间的协同操作需要考虑跨时钟的问题	(1) 数据不稳定, 微小的变化可能导致不同的树生成(2) 较难学习抽象概念	(1) 功能模块化、去耦, 高效可读、易于扩展(2) 使用多种节点类型来描述动作执行过程

Behavior tree based management and control architecture and scheduling method for networking service

RichHTML

PDF (PC)

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 9

References 15

Related Articles 1

Recommended Articles

Metrics

Comments