GAT-IL: A Service Function Chain Deployment Method Based on Graph Attention Network and Imitation Learning

FAN Qi-lin; NIU Yue; YIN Hao; WANG Tian-fu; LI Xiu-hua; HAO Jin-long

doi:10.12263/DZXB.20221180

您当前的位置：

首页 >

文章列表页 >

GAT-IL: A Service Function Chain Deployment Method Based on Graph Attention Network and Imitation Learning

PAPERS | 更新时间：2025-12-08

- GAT-IL: A Service Function Chain Deployment Method Based on Graph Attention Network and Imitation Learning
- ACTA ELECTRONICA SINICA Vol. 52, Issue 8, Pages: 2811-2823(2024)
- 作者机构：
  
  1.重庆大学大数据与软件学院,重庆 400044
  2.清华大学北京信息科学与技术国家研究中心,北京 100084
  3.中国科学技术大学计算机科学与技术学院,安徽合肥 230027
- 作者简介：
- 基金信息：
  
  National Natural Science Foundation of China(62102053;61972222;92067206;62072060)
- DOI：10.12263/DZXB.20221180
  CLC： TP393.0;
- Received：20 October 2022，
  
  Revised：2024-01-21，
  
  Published：25 August 2024
- 稿件说明：
移动端阅览
范琪琳, 牛岳, 尹浩, 等. GAT-IL：一种基于图注意力网络与模仿学习的服务功能链部署方法[J]. 电子学报, 2024, 52(08): 2811-2823.

FAN Qi-lin, NIU Yue, YIN Hao, et al. GAT-IL: A Service Function Chain Deployment Method Based on Graph Attention Network and Imitation Learning[J]. Acta Electronica Sinica, 2024, 52(08): 2811-2823.
范琪琳, 牛岳, 尹浩, 等. GAT-IL：一种基于图注意力网络与模仿学习的服务功能链部署方法[J]. 电子学报, 2024, 52(08): 2811-2823. DOI：10.12263/DZXB.20221180

FAN Qi-lin, NIU Yue, YIN Hao, et al. GAT-IL: A Service Function Chain Deployment Method Based on Graph Attention Network and Imitation Learning[J]. Acta Electronica Sinica, 2024, 52(08): 2811-2823. DOI：10.12263/DZXB.20221180

摘要

网络功能虚拟化通过将网络功能从专用硬件设备迁移到商用服务器上运行的软件中间盒中，简化了网络服务的配置和管理.在网络功能虚拟化的环境下，由一系列有序的虚拟网络功能组成的服务功能链正在成为承载网络服务的主流形式.将底层物理网络资源分配给服务功能链的需求称为服务功能链部署问题.对于基础设施提供商来说，在有限的资源条件下获得长期高回报是一个重要的挑战.本文形式化定义了服务功能链部署问题，提出了一种基于图注意力网络与模仿学习的服务功能链部署方法（Graph Attention Network and Imitation Learning，GAT-IL）.该方法使用图注意力网络评估每个物理服务器的放置潜力，通过蒙特卡洛树搜索方法给出专家示范，并采用模仿学习方法进行智能体的训练，融入集束搜索策略优化解空间.大量的实验结果表明，本文提出的GAT-IL方法在平均收益代价比和接受率的性能指标上均优于现有代表性算法.

Abstract

Network function virtualization simplifies the configuration and management of network services by migrating network functions from dedicated hardware devices to software middleboxes running on commercial servers. Under the environment of network function virtualization

the service function chain (SFC) composed of a series of ordered virtual network functions is becoming a mainstream alternative to host network services. The SFC deployment problem is to allocate the underlying physical network resources to the requirements of service function chains. It is challenging for infrastructure providers to obtain long-term high returns under limited resources. In this paper

we formally define the problem of SFC deployment and propose a novel method named graph attention network and imitation learning (GAT-IL) based on graph attention (GAT) network and imitation learning for SFC deployment. This method utilizes GAT to evaluate the potentials of each physical server

provides expert demonstrations through the Monte Carlo tree search algorithm

applies imitation learning to train the agent

and integrates the beam search strategy to optimize the solution space. Extensive experimental results show that the GAT-IL method proposed in this paper outperforms existing representative algorithms on performance metrics of average revenue-to-cost ratio and acceptance rate.

关键词

Keywords

references

LI D F , HONG P L , XUE K P , et al . Availability aware VNF deployment in datacenter through shared redundancy and multi-tenancy [J ] . IEEE Transactions on Network and Service Management , 2019 , 16 ( 4 ): 1651 - 1664 .

QU K G , ZHUANG W H , YE Q , et al . Dynamic flow migration for embedded services in SDN/NFV-enabled 5G core networks [J ] . IEEE Transactions on Communications , 2020 , 68 ( 4 ): 2394 - 2408 .

MIRJALILY G , LUO Z . Optimal network function virtualization and service function chaining: A survey [J ] . Chinese Journal of Electronics , 2018 , 27 ( 4 ): 704 - 717 .

COHEN R , LEWIN-EYTAN L , NAOR J S , et al . Near optimal placement of virtual network functions [C ] // Proceedings of the International Conference on Computer Communications . Piscataway : IEEE , 2015 : 1346 - 1354 .

LIU J Q , LI Y , ZHANG Y , et al . Improve service chaining performance with optimized middlebox placement [J ] . IEEE Transactions on Services Computing , 2015 , 10 ( 4 ): 560 - 573 .

LUIZELLI M C , COSTA CORDEIRO W L DA , BURIOL L S , et al . A fix-and-optimize approach for efficient and large scale virtual network function placement and chaining [J ] . Computer Communications , 2017 , 102 : 67 - 77 .

CHENG G Z , CHEN H C , HU H C , et al . Enabling network function combination via service chain instantiation [J ] . Computer Networks , 2015 , 92 : 396 - 407 .

YUE Y , CHENG B , LIU X , et al . Resource optimization and delay guarantee virtual network function placement for mapping SFC requests in cloud networks [J ] . IEEE Transactions on Network and Service Management , 2021 , 18 ( 2 ): 1508 - 1523 .

TANG P , LI F , ZHOU W , et al . Efficient auto-scaling approach in the telco cloud using self-learning algorithm [C ] // Proceedings of the Global Communications Conference . Piscataway : IEEE , 2015 : 1 - 6 .

XIAO Y K , ZHANG Q X , LIU F M , et al . NFVdeep: Adaptive online service function chain deployment with deep reinforcement learning [C ] // Proceedings of the International Symposium on Quality of Service . New York : ACM , 2019 : 1 - 10 .

WANG L , MAO W X , ZHAO J , et al . DDQP: A double deep Q-learning approach to online fault-tolerant SFC placement [J ] . IEEE Transactions on Network and Service Management , 2021 , 18 ( 1 ): 118 - 132 .

PAN P , FAN Q L , WANG S , et al . GCN-TD: A learning-based approach for service function chain deployment on the fly [C ] // Proceedings of the Global Communications Conference . Piscataway : IEEE , 2020 : 1 - 6 .

LI B Y , CHENG B , LIU X , et al . Joint resource optimization and delay-aware virtual network function migration in data center networks [J ] . IEEE Transactions on Network and Service Management , 2021 , 18 ( 3 ): 2960 - 2974 .

陈卓 , 冯钢 , 刘蓓 , 等 . 运营商网络中面向时延优化的服务功能链迁移重配置策略 [J ] . 电子学报 , 2018 , 46 ( 9 ): 2229 - 2237 .

CHEN Z , FENG G , LIU B , et al . Delay optimization oriented service function chain migration and re-deployment in operator network [J ] . Acta Electronica Sinica , 2018 , 46 ( 9 ): 2229 - 2237 . (in Chinese)

JIN P , FEI X , ZHANG Q , et al . Latency-aware VNF chain deployment with efficient resource reuse at network edge [C ] // Proceedings of the International Conference on Computer Communications . New York : ACM , 2020 : 267 - 276 .

LAGHRISSI A , TALEB T . A survey on the placement of virtual resources and virtual network functions [J ] . IEEE Communications Surveys & Tutorials , 2018 , 21 ( 2 ): 1409 - 1434 .

林嘉豪 , 章宗长 , 姜冲 , 等 . 基于生成对抗网络的模仿学习综述 [J ] . 计算机学报 , 2020 , 43 ( 2 ): 326 - 351 .

LIN J H , ZHANG Z C , JIANG C , et al . A survey of imitation learning based on generative adversarial nets [J ] . Chinese Journal of Computers , 2020 , 43 ( 2 ): 326 - 351 . (in Chinese)

KELLY M , SIDRANE C , DRIGGS-CAMPBELL K , et al . HG-DAgger: Interactive imitation learning with human experts [C ] // Proceedings of the International Conference on Robotics and Automation . Piscataway : IEEE , 2019 : 8077 - 8083 .

VELICKOVIC P , CUCURULL G , CASANOVA A , et al . Graph attention networks [J ] . Stat , 2017 , 1050 : 20 .

MAAS A L , HANNUN A Y , NG A Y . Rectifier nonlinearities improve neural network acoustic models [J ] . ICML Workshop on Deep Learning for Audio, Speech and Language Processing , 2013 , 30 ( 1 ): 3 .

HAERI S , TRAJKOVI Ć L . Virtual network embedding via Monte Carlo tree search [J ] . IEEE Transactions on Cybernetics , 2017 , 48 ( 2 ): 510 - 521 .

VODOPIVEC T , SAMOTHRAKIS S , STER B . On Monte Carlo tree search and reinforcement learning [J ] . Journal of Artificial Intelligence Research , 2017 , 60 : 881 - 936 .

MA X , DRIGGS-CAMPBELL K , ZHANG Z , et al . Monte-Carlo tree search for policy optimization [C ] // Proceedings of the 28th International Joint Conference on Artificial Intelligence . California : International Joint Conferences on Artificial Intelligence Organization , 2019 : 3116 - 3122 .

CAZENAVE T . Monte Carlo beam search [J ] . IEEE Transactions on Computational Intelligence and AI in Games , 2012 , 4 ( 1 ): 68 - 72 .

SIRONI C F , LIU J , WINANDS M H M . Self-adaptive Monte Carlo tree search in general game playing [J ] . IEEE Transactions on Games , 2018 , 12 ( 2 ): 132 - 144 .

KUO T W , LIOU B H , LIN K C J , et al . Deploying chains of virtual network functions: On the relation between link and server usage [J ] . IEEE/ACM Transactions on Networking , 2018 , 26 ( 4 ): 1562 - 1576 .

李茹杨 , 彭慧民 , 李仁刚 , 等 . 强化学习算法与应用综述 [J ] . 计算机系统应用 , 2020 , 29 : 13 - 25 .

LI R Y , PENG H M , LI R G , et al . Overview on algorithms and applications for reinforcement learning [J ] . Computer Systems & Applications , 2020 , 29 : 13 - 25 . (in Chinese)

WANG W Z , HONG P L , LEE D F , et al . Virtual network forwarding graph embedding based on tabu search [C ] // Proceedings of the 9th International Conference on Wireless Communications and Signal Processing . Piscataway : IEEE , 2017 : 1 - 6 .

FAN W B , XIAO F , CHEN X B , et al . Efficient virtual network embedding of cloud-based data center networks into optical networks [J ] . IEEE Transactions on Parallel and Distributed Systems , 2021 , 32 ( 11 ): 2793 - 2808 .

FU Z H , FAN Q L , ZHANG X , et al . Policy network assisted Monte Carlo tree search for intelligent service function chain deployment [C ] // Proceedings of the 20th International Conference on Trust, Security and Privacy in Computing and Communications . Piscataway : IEEE , 2021 : 1161 - 1168 .

YAO H , CHEN X , LI M , et al . A novel reinforcement learning algorithm for virtual network embedding [J ] . Neurocomputing , 2018 , 284 : 1 - 9 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Enhancing Multimodal Aspect-Based Sentiment Analysis with Adaptive Noise and Aspect Graph Association Learning

SCG-Detector: A Smart Contract Vulnerability Detection Method Based on Graph Attention Networks

A Node Classification Method Based on Graph Attention and Improved Transformer

Associations Prediction Algorithm of MiRNAs and Diseases Based on Heterogeneous Graph Attention Network

A Spatial-Temporal Dynamic Graph Attention Network Based Method for Sharing Travel Demand Prediction

Related Author

SONG Jian-hua

YANG Chao

ZHANG Yan

LIU Hui-jie

HUANG Chen

GU Xi-guo

WANG Zhi-wei

CHEN Xiang

Related Institution

School of Cyber Science and Technology, Hubei University

Hubei Province Project of Key Research Institute of Humanities and Social Sciences at Universities-RCIMPE

Hubei Key Laboratory of Big Data Intelligent Analysis and Application

Key Laboratory of Intelligent Sensing System and Security, Ministry of Education

School of Computer Science, Hubei University

⁰