The role of the Weibull distribution in modelling traffic in Internet access and backbone core networks

Abstract Over the years researchers have used Weibull distribution to model packet level Internet traffic behaviour without any physical or analytical justification other than its parameter flexibility and heavy-tailed behaviour. In this article we present an extensive data analysis of two-way traffic in various Internet access and backbone core links and show analytically why Internet traffic converges to Weibull distribution as traffic moves from access to core links. In addition we show the flexibility of Weibull distribution in capturing stochastic properties of Internet traffic at packet, flow and session levels in various access and backbone core links. An extensive literature survey with new developments in Internet traffic count data modelling has been presented. The contributions in this article establish the notion of the “Renewal of Renewal Theory in Internet Traffic Modelling”. This is the first study which presents a duplex analysis of all structural components of Internet traffic (packets, flows and sessions) at access and backbone core tiers of Internet. The results has been validated by using real traffic data fitness tests and trace driven queueing performance evaluation. The results of this article will help researchers use simple renewal processes as a better alternate to complex self-similar or modulated stochastic processes for modelling all structural components of Internet traffic at any time scale with physical justifications.

[1]  Georgios Smaragdakis,et al.  The growing complexity of content delivery networks: Challenges and implications for the Internet ecosystem , 2017 .

[2]  Kathleen S. Meier-Hellstern,et al.  TRAFFIC MODELS FOR ISDN DATA USERS: OFFICE AUTOMATION APPLICATION , 1991 .

[3]  David Hutchison,et al.  Internet traffic characterisation: Third-order statistics & higher-order spectra for precise traffic modelling , 2018, Comput. Networks.

[4]  Katerina Goseva-Popstojanova,et al.  Accounting for characteristics of session workloads: A study based on partly-open queue , 2012, 2012 IEEE International Conference on Communications (ICC).

[5]  Nikos Yannaros,et al.  Weibull renewal processes , 1994 .

[6]  R. Srikant,et al.  Impact of File Arrivals and Departures on Buffer Sizing in Core Routers , 2011, IEEE/ACM Transactions on Networking.

[7]  Pedro M. Mateo,et al.  A comparison between several adjustment models to simulated teletraffic data , 2007 .

[8]  Hao Jiang,et al.  Why is the internet traffic bursty in short time scales? , 2005, SIGMETRICS '05.

[9]  J. Gordon,et al.  Pareto process as a model of self-similar packet traffic , 1995, Proceedings of GLOBECOM '95.

[10]  Andrea Bianco,et al.  Web user session characterization via clustering techniques , 2005, GLOBECOM '05. IEEE Global Telecommunications Conference, 2005..

[11]  Susan L. Albin,et al.  Approximating a Point Process by a Renewal Process, II: Superposition Arrival Processes to Queues , 1984, Oper. Res..

[12]  Sally Floyd,et al.  Difficulties in simulating the internet , 2001, TNET.

[13]  Åke Arvidsson,et al.  Analysis of the Accuracy of Bursty Traffic Models , 1993 .

[14]  Riccardo Gusella,et al.  Characterizing the Variability of Arrival Processes with Indexes of Dispersion , 1991, IEEE J. Sel. Areas Commun..

[15]  Horst Rinne,et al.  The Weibull Distribution: A Handbook , 2008 .

[16]  Tayfur Altiok,et al.  Merging and splitting autocorrelated arrival processes and impact on queueing performance , 2008, Perform. Evaluation.

[17]  Andreas Willig,et al.  Fractal renewal process based analysis of emerging network traffic in access networks , 2016, 2016 26th International Telecommunication Networks and Applications Conference (ITNAC).

[18]  V. Moll,et al.  A generalized polygamma function , 2003, math/0305079.

[19]  Anja Feldmann,et al.  Characteristics of TCP Connection Arrivals , 2002 .

[20]  Jay M. Berger,et al.  A New Model for Error Clustering in Telephone Circuits , 1963, IBM J. Res. Dev..

[21]  V. Paxson,et al.  WHERE MATHEMATICS MEETS THE INTERNET , 1998 .

[22]  Gerhard Haßlinger,et al.  Content delivery and caching from a network provider's perspective , 2011, Comput. Networks.

[23]  Shingo Ata,et al.  Modeling of flows based on behavior of applications , 2012, The International Conference on Information Network 2012.

[24]  Michalis Faloutsos,et al.  Understanding the Limitations of Estimation Methods for Long-Range Dependence , 2006 .

[25]  Baris Balcioglu,et al.  A new renewal approximation for certain autocorrelated processes , 2008, Oper. Res. Lett..

[26]  Walter L. Smith,et al.  ON THE SUPERPOSITION OF RENEWAL PROCESSES , 1954 .

[27]  Åke Arvidsson,et al.  Performance Comparison of Bursty Traffic Models , 1991 .

[28]  Walter Willinger,et al.  Proof of a fundamental result in self-similar traffic modeling , 1997, CCRV.

[29]  Eric T. Bradlow,et al.  Count Models Based on Weibull Interarrival Times , 2008, 1307.5759.

[30]  Andreas Willig,et al.  The role of the Weibull distribution in Internet traffic modeling , 2013, Proceedings of the 2013 25th International Teletraffic Congress (ITC).

[31]  Walter Willinger,et al.  Self-similarity through high-variability: statistical analysis of Ethernet LAN traffic at the source level , 1997, TNET.

[32]  Pravin K. Trivedi,et al.  12 Count data models for financial data , 1996 .

[33]  Pablo Pavón-Mariño,et al.  Revisiting core traffic growth in the presence of expanding CDNs , 2019, Comput. Networks.

[34]  Richard G. Baraniuk,et al.  A Multifractal Wavelet Model with Application to Network Traffic , 1999, IEEE Trans. Inf. Theory.

[35]  D. McNickle,et al.  Towards a combined traffic modeling framework for access and core networks , 2012, Australasian Telecommunication Networks and Applications Conference (ATNAC) 2012.

[36]  J. Sethuraman,et al.  How Pooling Failure Data May Reverse Increasing Failure Rates , 1995 .

[37]  Jin Cao,et al.  Internet Traffic Tends Toward Poisson and Independent as the Load Increases , 2003 .

[38]  Walter Willinger,et al.  Experimental queueing analysis with long-range dependent packet traffic , 1996, TNET.

[39]  P. Kuehn,et al.  Approximate Analysis of General Queuing Networks by Decomposition , 1979, IEEE Trans. Commun..

[40]  Patrice Abry,et al.  Multifractality in TCP/IP traffic: the case against , 2005, Comput. Networks.

[41]  Jennifer Brown,et al.  Not all estimators are born equal: The empirical properties of some estimators of long memory , 2013, Math. Comput. Simul..

[42]  Zhi-Li Zhang,et al.  Small-time scaling behaviors of Internet backbone traffic: an empirical study , 2003, IEEE INFOCOM 2003. Twenty-second Annual Joint Conference of the IEEE Computer and Communications Societies (IEEE Cat. No.03CH37428).

[43]  Patrice Abry,et al.  Does fractal scaling at the IP level depend on TCP flow arrival processes? , 2002, IMW '02.

[44]  B.K. Ryu,et al.  Point process approaches to the modeling and analysis of self-similar traffic .I. Model construction , 1996, Proceedings of IEEE INFOCOM '96. Conference on Computer Communications.

[45]  Darryl Veitch,et al.  On the Role of Flows and Sessions in Internet Traffic Modeling: An Explorative Toy-Model , 2009, GLOBECOM 2009 - 2009 IEEE Global Telecommunications Conference.

[46]  Sally Floyd,et al.  Wide area traffic: the failure of Poisson modeling , 1995, TNET.

[47]  Richard Nelson,et al.  Libtrace: a packet capture and analysis library , 2012, CCRV.

[48]  Walter Willinger,et al.  On the self-similar nature of Ethernet traffic , 1993, SIGCOMM '93.

[49]  Jusak Jusak,et al.  WAVELET SPECTRUM FOR INVESTIGATING STATISTICAL CHARACTERISTICS OF UDP-BASED INTERNET TRAFFIC , 2012 .

[50]  S. Albin.  On Poisson Approximations for Superposition Arrival Processes in Queues , 1982 .

[51]  I. Kaj,et al.  Convergence of scaled renewal processes and a packet arrival model , 2003 .

[52]  Richard G. Clegg,et al.  A critical look at power law modelling of the Internet , 2009, Comput. Commun..

[53]  Richard Nelson,et al.  Measuring the accuracy of open-source payload-based traffic classifiers using popular Internet applications , 2013, 38th Annual IEEE Conference on Local Computer Networks - Workshops.

[54]  D. J. Daley,et al.  The Hurst Index of Long-Range Dependent Renewal Processes , 1999 .

[55]  Amir-Hossein Jahangir,et al.  An empirical study on TCP flow interarrival time distribution for normal and anomalous traffic , 2017, Int. J. Commun. Syst..

[56]  Rein Vesilo,et al.  Long-range dependent point processes and their Palm-Khinchin distributions , 2000 .

[57]  Jusak Jusak,et al.  Study of UDP-based Internet traffic: Long-range dependence characteristics , 2011, 2011 Australasian Telecommunication Networks and Applications Conference (ATNAC).

[58]  Azer Bestavros,et al.  Self-similarity in World Wide Web traffic: evidence and possible causes , 1997, TNET.

[59]  Teich,et al.  Fractal renewal processes generate 1/f noise. , 1993, Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics.

[60]  Antonio Nucci,et al.  Seeing the Difference in IP Traffic: Wireless Versus Wireline , 2006, Proceedings IEEE INFOCOM 2006. 25TH IEEE International Conference on Computer Communications.

[61]  K. Mitov,et al.  Superposition of renewal processes with heavy-tailed interarrival times , 2006 .

[62]  Frank Kelly,et al.  Notes on effective bandwidths , 1994 .

[63]  Fabio Ricciato,et al.  Revisiting an old friend: on the observability of the relation between long range dependence and heavy tail , 2010, Telecommun. Syst..

[64]  William Q. Meeker,et al.  Weibull Prediction Intervals for a Future Number of Failures , 2002, Technometrics.

[65]  Ward Whitt,et al.  Approximating a Point Process by a Renewal Process, I: Two Basic Methods , 1982, Oper. Res..

[66]  Michalis Faloutsos,et al.  A nonstationary Poisson view of Internet traffic , 2004, IEEE INFOCOM 2004.

[67]  Nick McKeown,et al.  Part I: buffer sizes for core routers , 2005, CCRV.

[68]  Wayne Nelson,et al.  Weibull prediction of a future number of failures , 2000 .

[69]  Andreas Willig,et al.  Global and local scaling analysis of link streams in access and backbone core networks , 2019, Comput. Networks.