VENDREDI 29 MAI 2015 APRES-MIDI

Session du Groupe de Travail DigiCosme

« Performances, QoS et gestion de ressources des réseaux sans fil dynamiques »

Lieu : Amphi 34, bâtiment 862 – NanoInnov

Avenue de la Vauve

91120 PALAISEAU

(GPS : N. 48°42,7575′, E. 02°11,5228′)

Entrée libre

14h00-15h00: Jim Roberts, IRT-SystemX – Titre : Multi-Resource Fairness: Objectives, Algorithms and Performance

Designing efficient and fair algorithms for sharing multiple resources between heterogeneous demands is becoming increasingly important. Applications include compute clusters shared by multi-task jobs and routers equipped with middleboxes shared by flows of different types. We show that the currently preferred objective of Dominant Resource Fairness (DRF) has a significantly less favorable efficiency-fairness tradeoff than alternatives like Proportional Fairness and our proposal, Bottleneck Max Fairness. We propose practical algorithms to realize these sharing objectives and evaluate their performance under a stochastic demand model. It is shown, in particular, that the strategyproofness property that motivated the choice of DRF for an assumed fixed set of jobs or flows, is largely irrelevant when demand is dynamic.

15h00-16h00 : Aline Carneiro Viana, INRIA – Titre: Mobile Data Traffic Modeling: Revealing Temporal Facets

Todays’ smart handheld devices allow heterogeneous free data gathering of human surrounding environment and networking usage patterns anytime and anywhere. Hence, an unprecedentedly large amount of human sensory data (i.e., the Big Data era) can be collected and processed: opening ways to connect people, technology, and business. This big data implies advanced knowledge of humans’ behaviour and interactions at a planetary scale and can help tackle networking challenges when used correctly. In this presentation, I will briefly talk about my most recent works where I leverage behavior of networking users (in terms of interactions, mobility, or content generation) to improve the performance of networking solutions designed for data offloading or dissemination.

I will then provide more details about our work on user content behavior modeling: Measurement-driven modeling of mobile data traffic usage of smartphone subscribers, which was based on the use of a large-scale dataset collected from a major 3G network in a big metropolitan area. Our main outcome is a synthetic measurement based mobile data traffic generator, capable of imitating traffic-related activity patterns of different categories of subscribers and time periods of a routinary normal day in their lives. Broadly, our observations bring important insights into network resource usage.

Prochain séminaire: vendredi 19 juin après-midi (orateurs: Marco Fiore, Alexandre Mouradian, Michel Marot)formation in a smart grid environment

Nous étudions la formation, dans le smart grid, de coalitions de prosumers cherchant à vendre de l’énergie sur le marché. Il est primordial pour l’opérateur du réseau électrique que les producteurs d’énergie soient à même de soutenir la demande du réseau à la fois en termes de stabilité et de production minimale. Nous avons ainsi construit un algorithme cherchant à former des coalitions répondant à ces deux pré-requis : un niveau minimum d’énergie produite et une production stable dans le temps. Ceci nous amène à rechercher des sources d’énergie non corrélées pour former les coalitions. Nous proposons donc un algorithme qui utilise des outils de la théorie des graphes comme les graphes de corrélation ou la percolation de cliques afin de former des coalitions répondant à ces contraintes complexes. Nous validons l’algorithme en le comparant avec un processus de regroupement aléatoire et nous montrons que notre algorithme se révèle plus performant non seulement en termes d’utilité globale, mais aussi en termes de robustesse face à des variations de production non prévues, dues par exemple à des changements de conditions climatiques.

15h00-16h00 : Daphné Tuncer, University College London – Titre: Network Resource Management and Control in Software Defined Networks

The Software-Defined Networking (SDN) paradigm has emerged as a promising solution to reduce the complexity of network management tasks through the creation of a unified control plane independent of specific vendor equipment. However, designing a SDN-based solution for network resource management raises several challenges as it should exhibit flexibility, scalability and adaptability. In this talk, we will present a new SDN-based management and control framework for fixed backbone networks, which provides support for both static and dynamic resource management applications. The proposed framework follows a hierarchical architecture and relies on a distributed layer to implement management and control functionality. We will present an approach to determine the allocation of managers and controllers in this layer, which relies on a set of input parameters. These can be tuned to take into account the requirements of both the network infrastructure and the management applications to implement. We will discuss the influence of these parameters on the configuration of the resulting management and control planes based on real network topologies and show how these can provide guidelines regarding the settings of the proposed placement algorithm.

Prochains séminaires:

Vendredi 19 Juin après-midi:

Orateurs: Marco Fiore – INRIA, Alexandre Mouradian – Université Paris Sud, Michel Marot – Institut Mines-Télécom