Robust and Reactive Traffic Engineering for Dynamic Traffic Demands

Abstract���The robustness and reliability of the Internet are highly dependent on efficient, stable connections and routing among provider networks. An analysis of real world trends in routing behavior across the Internet will have direct implications for the next generation of networking hardware, software and operational policies. This is why, observations of macroscopic traffic patterns provide insights into the legitimacy of the traffic, and how it affects network performance. Index Terms���Traffic anomalies, anomaly profile, routing

2008 16th European Signal Processing Conference, 2008

Optimal detection of unusual and significant changes in network Origin-Destination (OD) traffic volumes from simple link load measurements is considered in the paper. The ambient traffic, i.e. the OD traffic matrix corresponding to the non-anomalous network state, is unknown and it is considered here as a nuisance parameter because it can mask the anomalies. Since the OD traffic matrix is not recoverable from simple link load measurements, the anomaly detection is an ill-posed decision-making problem. The method proposed in this paper consists of finding a linear parsimonious model of ambient traffic (nuisance parameter) and detecting anomalies by using an invariant detection algorithm based on a separation of the measurement space into disjoint subspaces corresponding to normal and anomalous network traffic. The method's ability to detect anomalies is evaluated in real traffic from Abilene, a United States backbone network. The theoretically expected results are confirmed.

2008

Despite the bursty and highly volatile traffic, routing in the Internet today is optimised only on coarse time scales, as load-adaptive routing is known to induce performance deterioration by causing massive oscillations. We describe ReplEx, an universally applicable distributed algorithm for dynamic routing/traffic engineering, which is based on game theory. We show through extensive realistic simulations that ReplEx does not oscillate, and that it achieves performance gains comparable to traditional static traffic engineering. To obtain a better understanding of the traffic to be routed, we furthermore analyse global traffic matrices and search engine induced traffic, based on traffic measurements. Moreover we describe EaC, a universally applicable memory efficient algorithm that identify the most popular nodes in a search tree, which can serve to pin down main traffic contributors.

2006

One major challenge in communication networks is the problem of dynamically distributing load in the presence of bursty and hard to predict changes in traffic demands. Current traffic engineering operates on time scales of several hours which is too slow to react to phenomena like flash crowds or BGP reroutes. One possible solution is to use load sensitive routing. Yet, interacting routing decisions at short time scales can lead to oscillations, which has prevented load sensitive routing from being deployed since the early experiences in Arpanet. However, recent theoretical results have devised a game theoretical re-routing policy that provably avoids such oscillation and in addition can be shown to converge quickly. In this paper we present REPLEX, a distributed dynamic traffic engineering algorithm based on this policy. Exploiting the fact that most underlying routing protocols support multiple equal-cost routes to a destination, it dynamically changes the proportion of traffic th...

Procedia Computer Science, 2017

In your daily journeys, you are driving from a departure location to another destination location situated in the road network. There are a number of alternative routes that you could use. For each section of the network, we know the travel time required under normal traffic conditions to move from one endpoint to the other. What we do not know is the occurrence of incidents that slow down traffic and cause traffic congestion and therefore delays. This work considers the development of simple traffic model based on the semi-microscopic traffic assumption. The simulator based on the proposed model makes it possible to estimate the travel time on a road link. From variables, provided by simulator we can derive travel time index and therefore to characterize the traffic state of the urban transportation network. Another objective of this paper is to determine a robust itinerary in an urban transport network, taking into account the dynamic aspects of traffic congestion.

International Journal of Computer Applications, 2017

In digital era of technological revolution internet usage is rapidly increasing because of its need and usage to the society is unavoidable. As a result internet backbone facing traffic fueling issues in global perspective. To overcome this Traffic Engineering is an essential consideration in designing and operation of large internet backbone. Internet Traffic Engineering addresses the performance optimization of operational networks. Therefore constructing an effective optimistic algorithm for Traffic Engineering is a primary job to tackle the traffic fueling issues at internet backbones. This research article proposing an optimistic algorithmic approach for Congestion identification, avoidance and Traffic Distribution to Traffic Engineering mechanism better than existing conventional approach in execution.

2011

Traffic analysis and anomaly detection have been extensively used to characterize network utilization as well as to identify abnormal network traffic such as malicious attacks. However, so far, techniques for traffic analysis and anomaly detection have been carried out independently, relying on mechanisms and algorithms either in edge or in core networks alone. In this paper we propose the notion of multi-level network resilience, in order to provide a more robust traffic analysis and anomaly detection architecture, combining mechanisms and algorithms operating in a coordinated fashion both in the edge and in the core networks. This work is motivated by the potential complementarities between the research being developed at IIT Madras and Lancaster University. In this paper we describe the current work being developed at IIT Madras and Lancaster on traffic analysis and anomaly detection, and outline the principles of a multilevel resilience architecture.

Intra-domain Traffic Engineering (TE) aims to route around failures and congestion, and balance load in an ISP's network. Current TE relies on offline methods which use long term average traffic demands. It cannot react to traffic changes caused by BGP reroutes, diurnal traffic variations, attacks, or flash crowds. Further, current TE deals with network failures by pre-computing alternate routings for a limited set of failures. It may fail to prevent congestion when unanticipated or combination failures occur even though the network may have enough capacity to handle those failures. Since ISPs must ensure that each path in the network can cope with the peak load both during normal operation and under failures, current TE requires extensive over-provisioning. This paper presents TeXCP, an online distributed TE protocol that balances load in real-time, responding to actual traffic demands and failures. TeXCP uses multiple paths to deliver demands from an ingress to an egress router, adaptively moving traffic from overutilized to under-utilized paths. These adaptations are carefully designed such that, though done independently by each edge router based on local information, they balance load in the whole network without oscillations. We prove TeXCP is stable and show it is easy to implement. Our extensive simulations show that, for the same traffic demands, a network that uses TeXCP can support the same utilization and failure resilience as a network that uses traditional offline TE, but with a half or a third of the capacity.

The evaluation of dynamic Traffic Engineering (TE) algorithms is usually carried out using some specific network(s), traffic pattern(s) and traffic engineering objective(s). As the behavior of a TE algorithm is a consequence of the interactions between the network, the traffic demand and the algorithm itself, the relevance of TE may depend on several network design aspects. In this paper, we evaluate well-known TE algorithms using real-world and generated network topologies and traffic demands. By re-scaling observed traffic demands, we are able to observe the behavior of TE algorithms under a variety of situations, which may not be observable in reality. We identify distinct network load regimes that correspond to different behaviors of the TE algorithms. We also study the impact of several network design aspects, like network provisioning and redundancy, on the relevance of TE algorithms. We find that there are specific situations under which TE algorithms are useful. These situations depend highly on shortcomings in the network provisioning as well as on the availability of alternative paths in the network.

European Journal of Operational Research

We propose in this article an adaptive algorithm for optimal and robust guidance for the users of the road networks. The algorithm is based on the Stochastic On Time Arrival (SOTA) family of routing algorithms, which is appropriate for taking into account the variability of travel times through the road networks. The SOTA approach permits the derivation of the maximum cumulative probability distribution of the time arrival toward a given destination in the network. Those distributions allow the selection of the most reliable origin-destination paths under given travel time budgets. We investigate here the introduction of robustness against link and path failures in the criterion of the guidance strategy selection. Our algorithm takes into account the reliability of itinerary travel times, since it is based on a SOTA approach. In addition, the algorithm takes into account itinerary robustness, by favoring itineraries with possible and reliable alternative diversions, in case of link failures, with respect to itineraries without or with less reliable alternatives. We first analyze the algorithm in its static version, without considering the traffic dynamics, and show some interesting properties. We then combine the robust guidance algorithm with a dynamic traffic model by using the traffic simulator SUMO (Simulation of Urban Mobility), and illustrate its effectiveness in some dynamic scenarios.