计算机科学

计算机科学 ›› 2023, Vol. 50 ›› Issue (9): 331-336.doi: 10.11896/jsjkx.221000012

• 计算机网络 • 上一篇    下一篇

基于特征权重感知的VNF资源需求预测方法

王怀芹1, 骆健1,2, 王海艳1,2   

  1. 1 南京邮电大学计算机学院 南京 210023
    2 南京邮电大学江苏省大数据安全与智能处理重点实验室 南京 210023
  • 收稿日期:2022-10-07 修回日期:2023-03-17 出版日期:2023-09-15 发布日期:2023-09-01
  • 通讯作者: 王海艳(wanghy@njupt.edu)
  • 作者简介:(1220045225@njupt.edu.cn)
  • 基金资助:
    国家自然科学基金(62272243)

Feature Weight Perception-based Prediction of Virtual Network Function Resource Demands

WANG Huaiqin1, LUO Jian1,2, WANG Haiyan1,2   

  1. 1 School of Computer Science,Nanjing University of Post and Telecommunications,Nanjing 210023,China
    2 Jiangsu High Technology Research Key Laboratory for Wireless Sensor Networks,Nanjing University of Post and Telecommunications,Nanjing 210023,China
  • Received:2022-10-07 Revised:2023-03-17 Online:2023-09-15 Published:2023-09-01
  • About author:WANG Huaiqin,born in 1996,postgra-duate.Her main research interests includes service computing and deep lear-ning.
    WANG Haiyan,born in 1974,Ph.D,professor,is a senior member of China Computer Federation.Her main research interests include service computing,edge computing and big data intelligent processing technology.
  • Supported by:
    National Natural Science Foundation of China(62272243).

PDF (PC)

1697

摘要： 虚拟网络功能(Virtual Network Function,VNF)以服务功能链(Service Function Chain,SFC)的形式提供服务,能够满足不同服务的性能需求。由于网络具有动态性,为VNF实例分配固定资源会导致VNF实例的资源过多或者不足的问题。以往的研究对于VNF配置文件相关网络负载特征的重要性未做区分,因此,提出了一种基于特征权重感知的动态VNF资源需求预测方法。首先,使用ECANet学习VNF特征的权重值,以此来减少无用特征对模型预测结果的消极影响。其次,由于VNF配置文件数据集具有结构化特性,构建VNF资源预测模型时需要考虑以加强特征交互的方式来挖掘特征间深层的相互关系,提出使用深度特征交互网络(Deep Feature-Interactive Network,DIN) 增强网络负载特征与VNF性能特征之间的交互能力,提高模型预测精度。最后,在基准数据集上将所提方法与同类方法进行对比实验,发现其在预测的有效性与精确性上更具优势。

关键词: 资源预测, 服务功能链, 虚拟网络功能, 高效通道注意力网络, 特征交互

Abstract: Virtual network function(VNF) provides services in the form of service function chain(SFC) to meet the performance requirements of different services.Due to the dynamic nature of the network,allocating fixed resources to VNF instances will lead to excessive or insufficient resources for VNF instances.Previous studies have not distinguished the importance of network load characteristics related to VNF profiles.Therefore,a dynamic VNF resource demand prediction method based on feature weight perception is proposed.Firstly,ECANet is used to learn the weight values of VNF features,to reduce the negative impact of useless features on the model prediction results.Secondly,because the VNF profile data set has structural characteristics,when building the VNF resource prediction model,it is necessary to consider mining the deep interrelationship between features by strengthening feature interaction.It is proposed to use the deep feature interactive network(DIN) to enhance the interaction between network load features and VNF performance features,so as to improve the prediction accuracy of the model.Finally,compared with similar methods on the benchmark dataset,it is found that the proposed method has more advantages in the effectiveness and accuracy of prediction.

Key words: Resource prediction, Service function chain, Virtual network function, Efficient channel attention network, Feature interaction

中图分类号: 

  • TP391
[1]HERRERA J G,BOTERO J F.Resource allocation in NFV:A comprehensive survey[J].IEEE Transactions on Network and Service Management,2016,13(3):518-532.
[2]WU J W,JIANG L Y,LIU X J.VNF resource demand prediction method based on feature selection [J].Computer Application Research,2021,38(10):3131-3136,3142.
[3]KAFAFY M,FAHMY Y.Joint coding bit-rate and activity rate optimisation in wireless visual sensor networks[J].IET Communications,2020,14(19):3428-3439.
[4]LI X,TAN Y,MAREELS I,et al.Compatible formation set for uavs with visual sensing constraint[C]//2018 Annual American Control Conference(ACC).IEEE,2018:2497-2502.
[5]DIEYE M,AHVAR S,SAHOO J,et al.CPVNF:Cost-efficient proactive VNF placement and chaining for value-added services in content delivery networks[J].IEEE Transactions on Network and Service Management,2018,15(2):774-786.
[6]GAU R H.Optimal traffic engineering and placement of virtual machines in SDNs with service chaining[C]//2017 IEEE Conference on Network Softwarization(NetSoft).IEEE,2017:1-9.
[7]DRÄXLER S,SCHNEIDER S,KARL H.Scaling and placing bidirectional services with stateful virtual and physical network functions[C]//2018 4th IEEE Conference on Network Softwarization and Workshops(NetSoft).IEEE,2018:123-131.
[8]DRAXLER S,KARL H,MANN Z A.Joint optimization of sca-ling and placement of virtual network services[C]//2017 17th IEEE/ACM International Symposium on Cluster,Cloud and Grid Computing(CCGRID).IEEE,2017:365-370.
[9]DRÄXLER S,KARL H.SPRING:scaling,placement,and rou-ting of heterogeneous services with flexible structures[C]//2019 IEEE Conference on Network Softwarization(NetSoft).IEEE,2019:115-123.
[10]ERAMO V,MIUCCI E,AMMAR M,et al.An approach forservice function chain routing and virtual function network instance migration in network function virtualization architectures[J].IEEE/ACM Transactions on Networking,2017,25(4):2008-2025.
[11]KIM H G,LEE D Y,JEONG S Y,et al.Machine learning-based method for prediction of virtual network function resource demands[C]//2019 IEEE Conference on Network Softwarization(NetSoft).IEEE,2019:405-413.
[12]SCHNEIDER S,SATHEESCHANDRAN N P,PEUSTER M,et al.Machine learning for dynamic resource allocation in network function virtualization[C]//2020 6th IEEE Conference on Network Softwarization(NetSoft).IEEE,2020:122-130.
[13]COHEN R,LEWIN-EYTANL,NAOR J S,et al.Near optimalplacement of virtual network functions[C]//2015 IEEE Confe-rence on Computer Communications(INFOCOM).IEEE,2015:1346-1354.
[14]SALLAM G,JI B.Joint placement and allocation of virtual network functions with budget and capacity constraints[C]//IEEE INFOCOM 2019-IEEE Conference on Computer Communications.IEEE,2019:523-531.
[15]NTALAMPIRAS S,FIORE M.Forecasting mobile service de-mands for anticipatory MEC[C]//2018 IEEE 19th International Symposium on “A World of Wireless,Mobile and Multimedia Networks”(WoWMoM).IEEE,2018:14-19.
[16]MESTRES A,ALARCÓN E,CABELLOS A.A machine lear-ning-based approach for virtual network function modeling[C]//2018 IEEE Wireless Communications and Networking Conference Workshops(WCNCW).IEEE,2018:237-242.
[17]CHAUDHARI S,MITHAL V,POLATKAN G,et al.An attentive survey of attention models[J].ACM Transactions on Intelligent Systems and Technology(TIST),2021,12(5):1-32.
[18] WANG Q L,WU B G,ZHU P F,et al.ECA-Net:EfficientChannel Attention for Deep Convolutional Neural Networks[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR).IEEE,2020.
[19]HU J,SHEN L,SUN G.squeeze-and-excitation networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2018:7132-7141.
[20]HAN G,ZHANG M,WU W,et al.Improved U-Net based insulator image segmentation method based on attention mechanism[J].Energy Reports,2021,7:210-217.
[21]ZHOU C,LI D,WANG P,et al.ACR-Net:Attention Integrated and Cross-Spatial Feature Fused Rotation Network for Tubular Solder Joint Detection[J].IEEE Transactions on Instrumentation and Measurement,2021,70:1-12.
[22]QIN W,TANG J,LAO S.DeepFR:A trajectory prediction mo-del based on deep feature representation[J].Information Sciences,2022,604:226-248.
[23]JIANG B,LU Y,LU G,et al.Real noise image adjustment networks for saliency-aware stylistic color retouch[J].Knowledge-Based Systems,2022,242:108317.
[24]VAN R S,TAVERNIER W,COLLE D,et al.Profile-basedresource allocation for virtualized network functions[J].IEEE Transactions on Network and Service Management,2019,16(4):1374-1388.
[25]KIM H G,JEONG S Y,LEE D Y,et al.A deep learning approach to vnf resource prediction using correlation between vnfs[C]//2019 IEEE Conference on Network Softwarization(NetSoft).IEEE,2019:444-449.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发