Tractable effective bandwidths for end-to-end evaluation and fractional Brownian motion traffic

Effective bandwidth is a concept that has been developed for admission control as an indicator of the traffic load given to a network. However, that concept has been discussed mainly under a single node. That's because, while there are many studies on evaluation formulas for backlog at a single node, studies on end-to-end evaluation have been almost limited to ones by network calculus. In this paper, we develop an end-to-end backlog evaluation formula for a heterogeneous tandem network with cross traffic, as an extension of the one obtained in previous papers by the authors. The formula evaluates the asymptotic tail probability of the end-to-end backlog by the total traffic load at a bottleneck node, and here the traffic load of a flow is evaluated by a kind of effective bandwidth named tractable effective bandwidths (tEBW). The tEBW has a special property that makes the formula simple and tractable, and it is applicable to various types of input flows frequently used in performance analyses. Unfortunately, however, fractional Brownian motion (fBm) flows with long range dependency do not have any tEBW. For fBm flows, we show that another new tighter end-to-end backlog evaluation formula is effective, and using it we can show that, in a homogeneous tandem network with fBm cross traffic, the asymptotic tail probability of the end-to-end backlog is the same as that in the single node.

[1]  Yukio Takahashi,et al.  A Backlog Evaluation Formula for Admission Control Based on the Stochastic Network Calculus with Many Flows , 2011, IEICE Trans. Commun..

[2]  Jean-Yves Le Boudec,et al.  Network Calculus: A Theory of Deterministic Queuing Systems for the Internet , 2001 .

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

[4]  Amr Rizk,et al.  Statistical end-to-end performance bounds for networks under long memory FBM cross traffic , 2010, 2010 IEEE 18th International Workshop on Quality of Service (IWQoS).

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

[6]  Yong Liu,et al.  A calculus for stochastic QoS analysis , 2007, Perform. Evaluation.

[7]  Yukio Takahashi,et al.  A stochastic network calculus for many flows , 2009, 2009 21st International Teletraffic Congress.

[8]  Yukio Takahashi,et al.  Asymptotic end-to-end stochastic evaluation for tandem networks with many flows , 2009, VALUETOOLS.

[9]  Florin Ciucu,et al.  Delay Bounds in Communication Networks With Heavy-Tailed and Self-Similar Traffic , 2009, IEEE Transactions on Information Theory.

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

[11]  Florin Ciucu,et al.  A network service curve approach for the stochastic analysis of networks , 2005, SIGMETRICS '05.

[12]  Amir Dembo,et al.  Large Deviations Techniques and Applications , 1998 .

[13]  Cheng-Shang Chang,et al.  Performance guarantees in communication networks , 2000, Eur. Trans. Telecommun..

[14]  Yuming Jiang,et al.  Network calculus and queueing theory: two sides of one coin: invited paper , 2009, VALUETOOLS.

[15]  Chengzhi Li,et al.  A Network Calculus With Effective Bandwidth , 2007, IEEE/ACM Transactions on Networking.

[16]  Yong Liu,et al.  Stochastic Network Calculus , 2008 .