Steven Van den Berghe

An autonomic architecture for optimizing QoE in multimedia access networks

The recent emergence of multimedia services, such as Broadcast TV and Video on Demand over traditional twisted pair access networks, has complicated the network management in order to guarantee a decent Quality of Experience (QoE) for each user. The huge amount of services and the wide variety of service specifics require a QoE management on a per-user and per-service basis. This complexity can be tackled through the design of an autonomic QoE management architecture. In this article, the Knowledge Plane is presented as an autonomic layer that optimizes the QoE in multimedia access networks from the service originator to the user. It autonomously detects network problems, e.g. a congested link, bit errors on a link, etc. and determines an appropriate corrective action, e.g. switching to a lower bit rate video, adding an appropriate number of FEC packets, etc. The generic Knowledge Plane architecture is discussed, incorporating the triple design goal of an autonomic, generic and scalable architecture. The viability of an implementation using neural networks is investigated, by comparing it with a reasoner based on analytical equations. Performance results are presented of both reasoners in terms of both QoS and QoE metrics.

MPLS Recovery Mechanisms for IP-over-WDM Networks

Due to the fast increase of Internet traffic and the enormous bandwidth potential of all-optical transport networks based on wavelength division multiplexing, an IP-over-WDM network scenario is likely to be widespread in future communication networks. At the same time, IP networks are becoming more and more mission-critical. Hence, it is of paramount importance for IP-over-WDM networks to be able to recover quickly from frequently occurring network failures. This paper explains how multi-protocol label switching (both electrical and optical) recovery mechanisms can be important to reach that goal. Moreover, a novel MPLS recovery mechanism called fast topology-driven constrained-based rerouting is presented. Different MPLS recovery mechanisms are compared to each other. Special attention hereby goes to the additional capacity that is required to recover from frequently occurring failures.

Performance Evaluation of a Probabilistic Packet Filter Optimization Algorithm for High-Speed Network Monitoring

Because of the ever-increasing popularity of the Internet, network monitoring becomes very mission critical to guarantee the operation of IP networks, e.g. to detect network failures and stop intrusion attempts. A majority of these monitoring tasks require only a small subset of all passing packets, which share some common properties such as identical header fields or similar patterns in their data. Nowadays, next to the increasing network speed, much of these tasks become very complex. In order to capture only the useful packets, these applications need to evaluate a large set of expressions. In this paper, we present a platform independent filter and pattern matcher optimization algorithm, which reduces the required number of evaluated expressions. The performance of the algorithm will be validated both analytically and by means of a high-speed monitoring system.

Automated Generation of Knowledge Plane Components for Multimedia Access Networks

The management of Quality of Experience (QoE) in the access network is largely complicated by the wide range of offered services, the myriad of possible QoE restoring actions and the increasing heterogeneity of home network configurations. The Knowledge Plane is an autonomic framework for QoE management in the access network, aiming to provide QoE management on a per user and per service basis. The Knowledge Plane contains multiple problem solving components that determine the appropriate restoring actions. Due to the wide range of possible problems and the requirement of being adaptive to new services or restoring actions, it must be possible to easily add or adapt problem solving components. Currently, generating such a problem solving component takes a lot of time and needs manual tweaking. To enable an automated generation, we present the Knowledge Plane Compiler which takes a service management objective as input, stating available monitor inputs and relevant output actions and determines a suitable neural network based Knowledge Plane incorporating this objective. The architecture of the compiler is detailed and performance results are presented.

Packet filter optimization techniques for high-speed network monitoring

Effective network monitoring is vital for a growing number of control and management applications typically found in present-day networks. The ever-increasing link speeds and the complexity of monitoring applications’ needs have exposed severe limitations of existing monitoring techniques. A majority of the current monitoring tasks require only a small subset of all observed packets, which share some common properties such as identical header fields or similar patterns in their data. In order to capture only these useful packets, a large set of expressions needs to be evaluated. This evaluation should be done as efficiently as possible when monitoring multi-gigabit networks. To speed up this packet classification process, this article presents different packet filter optimization techniques. Complementary to existing approaches, we propose an adaptive optimization algorithm which dynamically reconfigures the filter expressions based on the currently observed traffic pattern. The performance of the algorithms is validated both analytically and by means of the implementation in a network monitoring framework. The various characteristics of the algorithms are investigated, including their performance in an operational network intrusion detection system.RésuméAujourd’hui, l’analyse du trafic des réseaux est essentielle pour la gestion de nombreuses applications. La croissance de la vitesse des réseaux et la complexité des besoins en analyse de trafic ont révélé des limitations sévères des méthodes et des outils existants. La plupart des tâches d’analyse n’exige qu’une partie de tous les paquets observés. Ces paquets partagent typiquement quelques propriétés comme des champs d’entêté identiques ou des séquences de caractères similaires dans leurs données. Pour capturer seulement ces paquets utiles, il faut évaluer une grande collection d’expressions. Cette évaluation doit être aussi performante que possible pour analyser des réseaux multi-gigabit. Cet article présente plusieurs techniques d’optimisation pour accélérer la procédure de classification des paquets. En complément des approches existantes, nous proposons un algorithme d’optimisation adaptif qui réorganise dynamiquement les expressions de filtrage à partir de l’observation du trafic. La performance des algorithmes est validée analytiquement et par une implémentation dans un système de surveillance du réseau. Les caractéristiques des algorithmes sont évaluées, y compris leur performance dans une application de détection d’intrusion opérationnelle.