In apparent contrast to the well-documented self-similar (i.e., monofractal) scaling behavior of measured LAN traffic, recent studies have suggested that measured TCP/IP and ATM WAN traffic exhibits more complex scaling behavior, consistent with multifractals. To bring multifractals into the realm of networking, this paper provides a simple construction based on cascades (also known as multiplicative processes) that is motivated by the protocol hierarchy of IP data networks. The cascade framework allows for a plausible physical explanation of the observed multifractal scaling behavior of data traffic and suggests that the underlying multiplicative structure is a traffic invariant for WAN traffic that co-exists with self-similarity. In particular, cascades allow us to refine the previously observed self-similar nature of data traffic to account for local irregularities in WAN traffic that are typically associated with networking mechanisms operating on small time scales, such as TCP flo...

This paper describes a framework for admission control for a packet-based network where the decisions are taken by edge devices or end-systems, rather than resources within the network. The decisions are based on the results of probe packets that the end-systems send through the network, and require only that resources apply a mark to packets in a way that is load dependent. One application example is the Internet, where marking information is fed back via an ECN bit, and we show howthis approach allows a rich QoS framework for ows or streams. Our approach allows networks to be explicitly analysed, and consequently engineered.

Recent measurements of network traffic have shown that self-similarity is an ubiquitous phenomenon present in both local area and wide area traffic traces. In previous work, we have shown a simple, robust application layer causal mechanism of traffic self-similarity, namely, the transfer of files in a network system where the file size distributions are heavy-tailed. In this paper, we study the effect of scale-invariant burstiness on network performance when the functionality of the transport layer and the nteraction of traffic sources sharing bounded network resources is incorporated. First, we show that transport layer mechanisms are important factors in translating the application layer causality into link traffic self-similarity. Network performance as captured by throughput, packet loss rate, and packet retransmission rate degrades gradually with increased heavy-tailedness while queueing delay, response time, and fairness deteriorate more drastically. The degree to which heavy-tailedness affects self-similarity is determined by how well congestion control is able to shape a source traffic into an on-average constant output stream while conserving information. Second, we show that increasing network resources such as link bandwidth and buffer capacity results in a superlinear improvement in performance. When large file transfers occur with nonnegligible probability, the incremental

Abstract—Destination-based forwarding in traditional IP routers has not been able to take full advantage of multiple paths that frequently exist in Internet Service Provider Networks. As a result, the networks may not operate efficiently, especially when the traffic patterns are dynamic. This paper describes a multipath adaptive traffic engineering mechanism, called MATE, which is targeted for switched networks such as MultiProtocol Label Switching (MPLS) networks. The main goal of MATE is to avoid network congestion by adaptively balancing the load among multiple paths based on measurement and analysis of path congestion. MATE adopts a minimalist approach in that intermediate nodes are not required to perform traffic engineering or measurements besides normal packet forwarding. Moreover, MATE does not impose any particular scheduling, buffer management, or a priori traffic characterization on the nodes. This paper presents an analytical model, derives a class of MATE algorithms, and proves their convergence. Several practical design techniques to implement MATE are described. Simulation results are provided to illustrate the efficacy of MATE under various network scenarios.

The Effect of Multiple Time Scales and Subexponentiality on the Behavior of a Broadband Network Multiplexer Predrag R. Jelenkovi'c The main theme of this dissertation is the evaluation of the capacity of broadband multimedia network multiplexers. This problem calls for the modeling of network traffic streams and the analysis of a network multiplexer that is loaded with the corresponding models. For modeling we focus on MPEG video traffic streams that are expected to be predominant in the traffic mixture of future multimedia networks. We experimentally demonstrate that real-time MPEG video traffic exhibits multiple time scale characteristics, as well as subexponential first and second order statistics. Then we construct a model of MPEG video that captures both of these characteristics and accurately predicts queueing behavior for a broad range of buffer and capacity sizes. Depending on whether a network multiplexer (loaded with MPEG) is strictly or weakly stable the dominant effect o...

In previous work [24], Fractal Point Processes (FPPs) have been proposed as novel tools for understanding, modeling and analyzing diverse types of self-similar traffic behavior. We apply the FPP models in the context of network traffic modeling and performance analysis. Two qualitatively different fractal data sets (Bellcore Ethernet traces) are characterized by FPP models. Comparison of model-driven and trace-driven queueing simulation results show that the matched models yield close agreement with the traces over a wide range of system parameters. We also show that under suitable conditions, the FPP models yield Gaussian processes. Queueing simulation shows that the FPP models can be computationally efficient alternatives for generating fractional Gaussian noise processes. Finally, we divide fractal traffic into two types, applicationlevel fractal traffic and network-level fractal traffic, and argue that each type has radically different implications for the design and control of fut...

In the Internet more and more network resource allocation problem arise at the connection level. Therefore it is essential to understand the nature of traffic at the connection level. This paper shows that the TCP connection arrival process shows self-similar behavior and that TCP connection interarrival times are statistically better modeled by distributions with heavy tails, especially the Weibull distribution, than traditional models. This implies that the connection arrival process is bursty. Given that a bursty arrival process can substantially degrade the performance of resource allocation algorithms we suggest the Weibull distribution as an alternative model for the analysis and simulations of such algorithms. 1 Introduction While packets are the basic unit of the Internet most operations by users involve more than one packet. The user experience dependents on the performance of the network on a set of packets. Sets of packets are starting to be used as basis for network...

In this paper, we develop a simple and powerful multiscale model for the synthesis of nonGaussian, long-range dependent (LRD) network traffic. Although wavelets effectively decorrelate LRD data, wavelet-based models have generally been restricted by a Gaussianity assumption that can be unrealistic for traffic. Using a multiplicative superstructure on top of the Haar wavelet transform, we exploit the decorrelating properties of wavelets while simultaneously capturing the positivity and "spikiness" of nonGaussian traffic. This leads to a swift O(N) algorithm for fitting and synthesizing N-point data sets. The resulting model belongs to the class of multifractal cascades, a set of processes with rich statistical properties. We elucidate our model's ability to capture the covariance structure of real data and then fit it to real traffic traces. Queueing experiments demonstrate the accuracy of the model for matching real data. Our results indicate that the nonGaussian nature of traffic has a significant effect on queuing.

Abstract not found

Analytical and empirical studies have shown that self�similar tra�c can have a detrimental impact on network performance including ampli�ed queuing delay and packet loss rate. Given the ubiquity of scale�invariant burstiness observed across diverse networking contexts � �nding e�ective tra�c control algorithms capable of detecting and managing self�similar tra�c has become an important problem. In this paper � we study congestion control algorithms for improving network performance�in particular � throughput�under self�similar tra�c conditions. Al� though scale�invariant burstiness implies the existence of concentrated periods of contention and idleness � the long�range dependence associated with self�similar traf� �c leaves open the possibility that correlation structure at larger time scales may be exploited for performance enhancement purposes. We construct a 2�level multiple time scale congestion control protocol that exer� cises congestion control concurrently across two time scales an order of magnitude apart. The �rst component�acting at the smaller time scale�is a generic linear increase�exponential decrease feedback congestion control that uses implicit predic� tion a�orded by feedback to a�ect rate control sensitive to changes in network state at 20�200ms time scales. The second component�acting at 2�5s time scales�uses explicit prediction to detect persistent shifts in overall network contention and uses this information to modulate the aggressiveness exhibited by the �rst component. We show that cooperative interaction between the two congestion control modules acting on information at di�erent time scales leads to improved performance vis� �a�vis the case when the large time scale component is absent. We show that the improvement factor increases with long�range dependence and we show that as the number of �ows engaging in multiple time scale congestion control �MTSC� increases � both fairness and e�ciency are preserved.