Learning end-to-end application QoS from openflow switch statistics

We use statistical learning to estimate end-to-end QoS metrics from device statistics, collected from a server cluster and an OpenFlow network. The results from our testbed, which runs a video-on-demand service and a key-value store, demonstrate that the learned models can estimate QoS metrics like frame rate or response time with errors bellow 10% for a given client. Interestingly, we find that service-level QoS metrics seem "encoded" in network statistics and it suffices to collect OpenFlow per port statistics to achieve accurate estimation at small overhead for data collection and model computation.

[1]  Vlado Menkovski,et al.  Online QoE prediction , 2010, 2010 Second International Workshop on Quality of Multimedia Experience (QoMEX).

[2]  Kaushik Dutta,et al.  Modeling virtualized applications using machine learning techniques , 2012, VEE '12.

[3]  José A. B. Fortes,et al.  On the Use of Machine Learning to Predict the Time and Resources Consumed by Applications , 2010, 2010 10th IEEE/ACM International Conference on Cluster, Cloud and Grid Computing.

[4]  Andy Liaw,et al.  Classification and Regression by randomForest , 2007 .

[5]  Mohamed Faten Zhani,et al.  DREAMS: Dynamic resource allocation for MapReduce with data skew , 2015, 2015 IFIP/IEEE International Symposium on Integrated Network Management (IM).

[6]  Seemab Latif,et al.  Handling intrusion and DDoS attacks in Software Defined Networks using machine learning techniques , 2014, 2014 National Software Engineering Conference.

[7]  Lingfen Sun,et al.  Learning models for video quality prediction over wireless local area network and universal mobile telecommunication system networks , 2010, IET Commun..

[8]  Helmut Hlavacs,et al.  Predicting web service levels during VM live migrations , 2011, 2011 5th International DMTF Academic Alliance Workshop on Systems and Virtualization Management: Standards and the Cloud (SVM).

[9]  Sanping Li.  Towards Ultra-High Speed Online Network Traffic Classification: Enhanced with Machine Learning Algorithms and OpenFlow Accelerators , 2013 .

[10]  Rolf Stadler,et al.  Predicting real-time service-level metrics from device statistics , 2015, 2015 IFIP/IEEE International Symposium on Integrated Network Management (IM).

[11]  Casimer DeCusatis,et al.  Predicting network attack patterns in SDN using machine learning approach , 2016, 2016 IEEE Conference on Network Function Virtualization and Software Defined Networks (NFV-SDN).

[12]  Rerngvit Yanggratoke,et al.  Data-driven Performance Prediction and Resource Allocation for Cloud Services , 2016 .

[13]  Luís Bernardo,et al.  Machine Learning in Software Defined Networks: Data collection and traffic classification , 2016, 2016 IEEE 24th International Conference on Network Protocols (ICNP).

[14]  Sidath Handurukande,et al.  Magneto approach to QoS monitoring , 2011, 12th IFIP/IEEE International Symposium on Integrated Network Management (IM 2011) and Workshops.

[15]  Rolf Stadler,et al.  Predicting service metrics for cluster-based services using real-time analytics , 2015, 2015 11th International Conference on Network and Service Management (CNSM).

[16]  Antonio Liotta,et al.  Predicting quality of experience in multimedia streaming , 2009, MoMM.

[17]  川野 秀一.  An Introduction to Statistical Learning (with Applications in R), Gareth James,Daniela Witten,Trevor Hastie and Robert Tibshirani著, Springer, 2013年8月, 430pp., 価格 59.99〓, ISBN 978-1-4614-7137-0 , 2014 .

[18]  Michael I. Jordan,et al.  Statistical Machine Learning Makes Automatic Control Practical for Internet Datacenters , 2009, HotCloud.

[19]  Yin Zhang,et al.  Q-score: proactive service quality assessment in a large IPTV system , 2011, IMC '11.

[20]  Bo Hong,et al.  Managing flash crowds on the Internet , 2003, 11th IEEE/ACM International Symposium on Modeling, Analysis and Simulation of Computer Telecommunications Systems, 2003. MASCOTS 2003..

[21]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.