面向多模态网络业务切片的虚拟网络功能资源容量智能预测方法

doi:10.11959/j.issn.1000-436x.2022098

摘要/Abstract

摘要：

目的：随着多模态网络等新型网络结构的出现和发展，对于网络资源容量的需求也日趋多样化。调整网络切片上承载的虚拟网络功能（VNF）的分布，进行按需、动态、高效的资源容量匹配十分困难。本文利用数据驱动的VNF资源容量预测方法，探索网络切片上承载VNF的预部署。

方法：利用数据驱动的思想，采用一种基于时空特征提取的VNF资源容量预测方法，通过资源容量预测为即将到来的网络切片需求进行VNF的预部署。首先，对用于预测的数据流时间序列进行两段加权处理，然后把处理后的时间序列及其依赖的空间拓扑信息输入网络模型中，进行时空特征提取。对于空间特征的提取，通过给定一个邻接矩阵和一个特征矩阵，利用图卷积网络在傅里叶域中重组时间序列的空间分布特征。对于时间特征的提取，经由门控循环单元，通过单元之间的信息传递感知输入数据的时序依赖关系。然后，基于数据流序列与VNF实例数量的映射关系，由前馈神经网络进行数据维度转换并最终输出VNF资源需求预测视图。

结果：从实验结果可以看出，该方法具有的预测性能主要体现在以下几个方面：①该方法的预测精度稳定。首要原因在于该方法在结构上采用了合理的时空特征提取结构，该结构能够有效地同时处理具有非欧几里得特征的空间结构和具有前后依赖性的时间序列特征。次要原因在于该方法对于输入数据流序列进行加权处理的预处理过程有效规避了网络中突发流带来的特征突变，能够真实的反映数据流以及切片容量需求的变化趋势，因而获得了稳定的预测效果。②该方法具有时空预测能力。说明在经过大量数据训练后，该方法的空间特征提取层能够迅速对网络的拓扑关系和数据流分布进行计算，时间特征提取层则能够根据节点间数据流之间潜在的相关性辅助预判数据流可能出现的突然变化，两个特征提取层协调运转，获得准确的时空预测结果。③该方法具有数据流预测结果转换能力。说明该方法通过结合数据流波动趋势与VNF实例数量变化趋势之间的映射关系，通过数据维度的转换，高效地实现了数据流预测与网络切片资源容量预测的转化。

结论：多模态网络借助人工智能技术的蓬勃发展，为切片上的虚拟节点赋予计算、存储和传输能力，已经能够在自主感知数据吞吐和自主预测节点VNF的资源需求的基础上，通过业务需求数据的自适应流动，实现网络资源与用户需求的按需匹配。本文借助机器学习算法提出的用于多种模态网络场景中VNF资源容量需求预测方法VNFPre，可以判断网络切片在未来的VNF资源容量需求，为网络切片承载的VNF的放置和映射提供先验信息。

关键词: 多模态网络, 虚拟网络功能, 资源容量, 时空特征提取

Abstract:

Objectives:With the emergence and development of new network structures such as polymorphic network,the demand for network resource capacity is becoming more and more diverse.It is very difficult to adjust the distribution of virtual network functions (VNFs) carried on network slices for on-demand,dynamic and efficient resource capacity matching.This paper uses a data-driven VNF resource capacity prediction method to explore the pre-deployment of VNFs carried on network slices.

Methods: Using the data-driven idea, a VNF resource capacity prediction method based on spatiotemporal feature extraction was adopted to pre-deploy VNFs for the upcoming slicing demand through resource capacity prediction. First, the time series of data stream used for prediction is subjected to two-stage weighting processing,and then the processed time series and its dependent spatial topology information are input into the network model for spatiotemporal feature extraction. For the extraction of spatial features, by given an adjacency matrix and a feature matrix,graph convolutional network is used to reorganize the spatial distribution features of time series in the Fourier domain. For the extraction of temporal features, the temporal dependencies of the input data are perceived through the information transfer between the units via gated recurrent units.Then,based on the mapping relationship between the data flow sequence and the number of VNF instances, the feedforward neural network performs data dimension transformation and finally outputs the VNF resource demand prediction results.

Results: From the experimental results, the prediction performance of this method is mainly reflected in the following aspects:1.The prediction accuracy of this method is stable.The primary reason is that the method adopts a reasonable spatiotemporal feature extraction structure, which can effectively deal with both the spatial structure with non-Euclidean features and the time series features with contextual dependencies.The secondary reason is that the weighted preprocessing process of the input data stream sequence in this method effectively avoids the characteristic mutation caused by the burst flow in the network,and can truly reflect the changing trend of the data stream and slicing capacity requirements, thus obtaining stable prediction effect. 2. The method has the ability of spatiotemporal prediction. It shows that after a large amount of data training, the spatial feature extraction layer of the method can quickly calculate the topology relationship and data flow distribution of the network, and the temporal feature extraction layer can assist in predicting the sudden changes may occur in the data flow according to the potential correlation between data flows between nodes. The two feature extraction layers work in coordination to obtain accurate spatiotemporal prediction results.3.The method has the ability to convert data flow prediction results. It shows that this method can combine the mapping relationship between the fluctuation trend of data flow and the change trend of the number of VNF instances, and efficiently realizes the conversion of data flow prediction and network slice capacity prediction through the transformation of data dimension.

Conclusions: With the vigorous development of artificial intelligence technology, polymorphic network has given computing, storage and transmission capabilities to virtual nodes on network slices,and has been able to realize the dual improvement of network resource utilization and user experience through the adaptive flow of business demand data on the basis of autonomously sensing data throughput and autonomously predicting the resource requirements of VNFs on nodes. With the help of machine learning algorithms, the VNF resource capacity demand prediction method VNFPre proposed for polymorphic network scenarios,it can judge the future VNF resource capacity demand of network slices, and provide a priori information for the placement and mapping of VNFs carried by network slices.

Key words: polymorphic network, virtual network function, resource capacity, spatial-temporal feature extraction

中图分类号:

TP393

兰巨龙, 朱棣, 李丹. 面向多模态网络业务切片的虚拟网络功能资源容量智能预测方法[J]. 通信学报, 2022, 43(6): 143-155.

Julong LAN, Di ZHU, Dan LI. Intelligent prediction method of virtual network function resource capacity for polymorphic network service slicing[J]. Journal on Communications, 2022, 43(6): 143-155.

图/表 17

图1

图2

图3

图4

图5

图6

图7

表1

图8

图9

图10

图11

图12

图13

图14

图15

图16

参考文献 25

[1]	兰巨龙, 程东年, 胡宇翔 . 可重构信息通信基础网络体系研究[J]. 通信学报, 2014,35(1): 128-139.
	LAN J L , CHENG D N , HU Y X . Research on reconfigurable information communication basal network architecture[J]. Journal on Communications, 2014,35(1): 128-139.
[2]	胡宇翔, 伊鹏, 孙鹏浩 ,等. 全维可定义的多模态智慧网络体系研究[J]. 通信学报, 2019,40(8): 1-12.
	HU Y X , YI P , SUN P H ,et al. Research on the full-dimensional defined polymorphic smart network[J]. Journal on Communications, 2019,40(8): 1-12.
[3]	NGMN Alliance. Description of network slicing concept[EB].(2017-09-11)[2022-01-04]
[4]	AFOLABI I , BAGAA M , TALEB T ,et al. End-to-end network slicing enabled through network function virtualization[C]// Proceedings of 2017 IEEE Conference on Standards for Communications and Networking. Piscataway:IEEE Press, 2017: 30-35.
[5]	ZAMAN Z , RAHMAN S , NAZNIN M . Novel approaches for VNF requirement prediction using DNN and LSTM[C]// 2019 IEEE Global Communications Conference (GLOBECOM). Piscataway:IEEE Press, 2019: 1-6.
[6]	QIU F , ZHANG B , GUO J . A deep learning approach for VM workload prediction in the cloud[C]// Proceedings of 2016 17th IEEE/ACIS International Conference on Software Engineering,Artificial Intelligence,Networking and Parallel/Distributed Computing (SNPD). Piscataway:IEEE Press, 2016: 319-324.
[7]	唐伦, 周钰, 杨友超 ,等. 5G网络切片场景中基于预测的虚拟网络功能动态部署算法[J]. 电子与信息学报, 2019,41(9): 2071-2078.
	TANG L , ZHOU Y , YANG Y C ,et al. Virtual network function dynamic deployment algorithm based on prediction for 5G network slicing[J]. Journal of Electronics ＆ Information Technology, 2019,41(9): 2071-2078.
[8]	TANG H , ZHOU D , CHEN D . Dynamic network function instance scaling based on traffic forecasting and VNF placement in operator data centers[J]. IEEE Transactions on Parallel and Distributed Systems, 2019,30(3): 530-543.
[9]	ALAWE I , HADJADJ-AOUL Y KSENTINI A ,et al. Smart scaling of the 5G core network:an RNN-based approach[C]// Proceedings of 2018 IEEE Global Communications Conference. Piscataway:IEEE Press, 2018: 1-6.
[10]	THAKKAR H K , DEHURY C K , SAHOO P K . MUVINE:multi-stage virtual network embedding in cloud data centers using reinforcement learning-based predictions[J]. IEEE Journal on Selected Areas in Communications, 2020,38(6): 1058-1074.
[11]	PATEL Y S , VERMA D , MISRA R . Deep learning based resource allocation for auto-scaling VNFs[C]// Proceedings of 2019 IEEE International Conference on Advanced Networks and Telecommunications Systems. Piscataway:IEEE Press, 2019: 1-6.
[12]	SCALINGI A , ESPOSITO F , MUHAMMAD W ,et al. Scalable provisioning of virtual network functions via supervised learning[C]// Proceedings of 2019 IEEE Conference on Network Softwarization (NetSoft). Piscataway:IEEE Press, 2019: 423-431.
[13]	TONG R M , XU S Y , HU B ,et al. VNF dynamic scaling and deployment algorithm based on traffic prediction[C]// Proceedings of 2020 International Wireless Communications and Mobile Computing (IWCMC). Piscataway:IEEE Press, 2020: 789-794.
[14]	TAO J , LU Z , CHEN Y ,et al. Adaptive VNF scaling approach with proactive traffic prediction in NFV-enabled clouds[C]// ACM Turing Award Celebration Conference. New York:ACM Press, 2021: 166-172.
[15]	MIJUMBI R , HASIJA S , DAVY S ,et al. Topology-aware prediction of virtual network function resource requirements[J]. IEEE Transactions on Network and Service Management, 2017,14(1): 106-120.
[16]	JALODIA N , HENNA S , DAVY A . Deep reinforcement learning for topology-aware VNF resource prediction in NFV environments[C]// Proceedings of 2019 IEEE Conference on Network Function Virtualization and Software Defined Networks. Piscataway:IEEE Press, 2019: 1-5.
[17]	KIM H G , JEONG S Y , LEE D Y ,et al. A deep learning approach to VNF resource prediction using correlation between VNFs[C]// Proceedings of 2019 IEEE Conference on Network Softwarization (NetSoft). Piscataway:IEEE Press, 2019: 444-449.
[18]	ZHUANG Z R , WANG J Y , QI Q ,et al. Toward greater intelligence in route planning:a graph-aware deep learning approach[J]. IEEE Systems Journal, 2020,14(2): 1658-1669.
[19]	SHEN Y F , SHI Y M , ZHANG J ,et al. Graph neural networks for scalable radio resource management:architecture design and theoretical analysis[J]. IEEE Journal on Selected Areas in Communications, 2021,39(1): 101-115.
[20]	FOUKAS X , PATOUNAS G , ELMOKASHFI A ,et al. Network slicing in 5G:survey and challenges[J]. IEEE Communications Magazine, 2017,55(5): 94-100.
[21]	THOMAS R W , FRIEND D H , DASILVA L A ,et al. Cognitive networks:adaptation and learning to achieve end-to-end performance objectives[J]. IEEE Communications Magazine, 2006,44(12): 51-57.
[22]	KIPF T N , WELLING M . Semi-supervised classification with graph convolutional networks[J]. arXiv Preprint,arXiv:1609.02907, 2016.
[23]	UHLIG S , QUOITIN B , LEPROPRE J ,et al. Providing public intradomain traffic matrices to the research community[J]. ACM SIG COMM Computer Communication Review, 2006,36(1): 83-86.
[24]	武静雯, 江凌云, 刘祥军 . 基于特征选择的 VNF 资源需求预测方法[J]. 计算机应用研究, 2021,38(10): 3131-3136,3142.
	WU J W , JIANG L Y , LIU X J . VNF resource demand forecast method based on feature selection[J]. Application Research of Computers, 2021,38(10): 3131-3136,3142.
[25]	WANG B , LUO X Y , ZHANG F B ,et al. Graph-based deep modeling and real time forecasting of sparse spatio-temporal data[J]. arXiv Preprint,arXiv:1804.00684, 2018.

方法	R²	RMSE
VNFPre	0.973 0	29.142 0
VNFPre-input	0.897 5	30.212 5
CNN-LSTM	0.913 3	30.137 6
LSTM	0.824 3	32.082 1