Performance modeling and network management for self-similar traffic

The main goal behind the creation of Asynchronous Transfer Mode network is to use bandwidth more efficiently by time-multiplexing bursty sources while still being able to offer Quality of Service guarantees. In order to meet this goal, it is necessary to have analytical models that can accurately replicate the behavior of real ATM networks and traffic. This work is concerned with improving our theoretical understanding of network traffic, developing more accurate and useful performance models, and applying the information gathered from those models to develop more effective network management techniques. Since the discovery of the self-similar nature of network traffic, researchers were able to derive new traffic models that are better able to mimic the long-range dependence phenomenon exhibited by real network traffic. In this work, we investigate new analytical tools in order to handle an ATM queueing system driven by a self-similar process and to verify the impact of these new traffic assumptions on current ATM protocols. We propose a new traffic characterization, we desire a new framework capable of computing bandwidth and buffer requirements in ATM networks driven by a self-similar process. The novelty in our approach is the use of a probabilitics envelope process in order to compute tail probabilities of the buffer occupancy of an ATM multiplexer. We derive a new probabilistic delay bound and show that it (i) agrees with delay experimented by real network traffic and (ii) matches the results obtained by large deviation theory. Furthermore, we extend this framework in order to (i) estimate end-to-end delay bounds in tandem queueing networks driven by a self-similar process, (ii) to queueing systems with a non-deterministic service time distribution, and (iii) to find tail probabilities of the buffer occupancy of an ATM multiplexer driven by heterogeneous long-range dependent sources. At the best of our knowledge, this is the only framework capable of computing end-to-end for tandum-queue networks driven by aggregate heterogeneous long-range dependent sources.

[1]  Anwar Elwalid,et al.  The importance of long-range dependence of VBR video traffic in ATM traffic engineering: myths and realities , 1996, SIGCOMM 1996.

[2]  M. Taqqu,et al.  Using Renewal Processes to Generate Long-Range Dependence and High Variability , 1986 .

[3]  A. I. McLeod,et al.  Preservation of the rescaled adjusted range: 1. A reassessment of the Hurst Phenomenon , 1978 .

[4]  Rene L. Cruz,et al.  A calculus for network delay, Part I: Network elements in isolation , 1991, IEEE Trans. Inf. Theory.

[5]  Vern Paxson,et al.  Fast approximation of self-similar network traffic , 1995, SIGCOMM 1995.

[6]  J. Hosking.  Modeling persistence in hydrological time series using fractional differencing , 1984 .

[7]  B. Mandelbrot.  A Fast Fractional Gaussian Noise Generator , 1971 .

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

[9]  P. Pruthi,et al.  Heavy-tailed on/off source behavior and self-similar traffic , 1995, Proceedings IEEE International Conference on Communications ICC '95.

[10]  Ilkka Norros,et al.  A storage model with self-similar input , 1994, Queueing Syst. Theory Appl..

[11]  Moshe Zukerman,et al.  Fractal traffic: measurements, modelling and performance evaluation , 1995, Proceedings of INFOCOM'95.

[12]  Nicolas D. Georganas,et al.  Analysis of an ATM buffer with self-similar ("fractal") input traffic , 1995, Proceedings of INFOCOM'95.

[13]  Michael Devetsikiotis,et al.  Fast simulation for self-similar traffic in ATM networks , 1995, Proceedings IEEE International Conference on Communications ICC '95.

[14]  John A. Silvester,et al.  Time scale analysis of an ATM queueing system with long-range dependent traffic , 1997, Proceedings of INFOCOM '97.

[15]  Ying-Dar Lin,et al.  Virtual path management in ATM networks , 1996, Proceedings of ICC/SUPERCOMM '96 - International Conference on Communications.

[16]  Ray Jain,et al.  The art of computer systems performance analysis - techniques for experimental design, measurement, simulation, and modeling , 1991, Wiley professional computing.

[17]  On the accuracy of approximating loss probabilities in finite queues by probabilities to exceed queue levels in infinite queues , 1998, IEEE GLOBECOM 1998 (Cat. NO. 98CH36250).

[18]  D. Applebaum.  Stable non-Gaussian random processes , 1995, The Mathematical Gazette.

[19]  K. Knight.  Stable Non-Gaussian Random Processes Gennady Samorodnitsky and Murad S. Taqqu Chapman and Hall, 1994 , 1997, Econometric Theory.

[20]  A. Erramilli,et al.  Self-similar traffic generation: the random midpoint displacement algorithm and its properties , 1995, Proceedings IEEE International Conference on Communications ICC '95.

[21]  B. Mandelbrot,et al.  Fractional Brownian Motions, Fractional Noises and Applications , 1968 .

[22]  G. Mayor,et al.  The multi-level leaky bucket mechanism , 1995, Proceedings of Fourth International Conference on Computer Communications and Networks - IC3N'95.

[23]  San-qi Li,et al.  Spectral analysis of packet loss rate at a statistical multiplexer for multimedia services , 1994, TNET.

[24]  San-qi Li,et al.  Queue response to input correlation functions: discrete spectral analysis , 1993, TNET.

[25]  B. Mandelbrot.  Long-Run Linearity, Locally Gaussian Process, H-Spectra and Infinite Variances , 1969 .

[26]  Michael Devetsikiotis,et al.  Modeling and simulation of self-similar variable bit rate compressed video: a unified approach , 1995, SIGCOMM '95.

[27]  Mark W. Garrett,et al.  Modeling and generation of self-similar vbr video traffic , 1994, SIGCOMM 1994.

[28]  John A. Silvester,et al.  Simulation of an ATM queueing system with a fractional Brownian noise arrival process , 1996, Proceedings of ICC/SUPERCOMM '96 - International Conference on Communications.

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

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

[31]  G. K. Young,et al.  Practical application of fractional Brownian Motion and noise to synthetic hydrology , 1973 .

[32]  Abhay Parekh,et al.  A generalized processor sharing approach to flow control in integrated services networks: the single-node case , 1993, TNET.

[33]  Erwin P. Rathgeb.  Modeling and Performance Comparison of Policing Mechanisms for ATM Networks , 1991, IEEE J. Sel. Areas Commun..

[34]  Gustavo de Veciana,et al.  On the relevance of time scales in performance oriented traffic characterizations , 1996, Proceedings of IEEE INFOCOM '96. Conference on Computer Communications.

[35]  Cheng-Shang Chang,et al.  Stability, queue length, and delay of deterministic and stochastic queueing networks , 1994, IEEE Trans. Autom. Control..

[36]  Moshe Zukerman,et al.  Performance of a single server queue with self similar input , 1995, Proceedings IEEE International Conference on Communications ICC '95.

[37]  Jan Beran,et al.  Statistics for long-memory processes , 1994 .

[38]  Mischa Schwartz,et al.  Broadband integrated networks , 1996 .

[39]  Armand M. Makowski,et al.  Tail probabilities for a multiplexer with self-similar traffic , 1996, Proceedings of IEEE INFOCOM '96. Conference on Computer Communications.

[40]  I. Norros,et al.  The management of large flows of connectionless traffic on the basis of self-similar modeling , 1995, Proceedings IEEE International Conference on Communications ICC '95.

[41]  N. Duffield,et al.  Predicting quality of service for traffic with long-range fluctuations , 1995, Proceedings IEEE International Conference on Communications ICC '95.