Exposé invité de Alain Tchana (INP Toulouse) le 02/02/2018 à 10h30 dans l'amphi du Labri

Bonjour à tous,

Alain Tchana de l'INPT fera un exposé le vendredi 02/02/2018 à 10h30 dans l'amphi du Labri sur:

Energy saving and performance improvement in virtualized datacenters

Numerous companies rely on cloud infrastructures for hosting their services because of their
attractive costs. Electricity consumption is a major challenge in such infrastructures, it
represents around 50% to 70% of the expenditures. It doubles every five years due to
explosive growth in both the number and density of datacenters, according to the last Berkeley Lab report. So, this is also a critical issue for the planet because electricity consumption results in CO2 emission. Existing approaches, based on virtual machine (VM) packing and dynamic server power scaling, have reached their limit because of the server-centric architecture of datacenters on the one hand and the server bounded limitation of VM monitors on the other hand. The new trend to go further consists in building future datacenters as giant computers, called disaggregated datacenters. I will present my major contributions in this direction.

Since September 2013, Alain Tchana is Associate Professor - HDR at ENSEEIHT-INPT. His Ph.D. thesis was on research topics related to autonomic administration systems. The latter are middleware services which allow administration tasks to be automated (self-deployment, self-repair, self-*). In particular, he worked on the autonomic administration of cloud infrastructures. Since he was recruited as Associate Professor, he moved to the virtualization and operating system domains, (the good layers for addressing issues in their generality). The next step in his research path is the exploration of the future computer architectures for cloud infrastructures.

Séminaire de Hermann Hellwagner sur le "Dynamic Adaptive Streaming over Named Data Networking" le 21/11/2017 de 10h à 11h en salle 076 du Labri

Dynamic Adaptive Streaming over Named Data Networking

Hermann Hellwagner (joint work with Daniel Posch and Benjamin Rainer)
Multimedia Communication Research Group
Institute of Information Technology
Alpen-Adria-Universität Klagenfurt (AAU), Austria

Modern digital video portals and streaming services typically store multiple, differently encoded variants of the available videos and employ adaptation techniques in order to adapt the volume and quality of the transmitted video data, in response to, e.g., dynamically varying network throughput conditions or specific constraints of end devices. In the past decade, client-driven adaptive video delivery techniques have become prevailing in practice, e.g., the standardized Dynamic Adaptive Streaming over HTTP(DASH) scheme or its proprietary variants. The Multimedia Communication research group at AAU has for several years explored DASH-type schemes. The spin-off company BITMOVIN successfully productized the software and extended it substantially. 

In this talk, though, we will present a second line of research that developed out of the first one, namely how to realize Dynamic Adaptive Streaming(DAS) over Named Data Networking (NDN), a specific and promising architecture variant of Information-Centric Networking (ICN) investigated as part of Future Internet initiatives. The basic NDN architecture was designed several years ago by Jacobson et al. at PARC to address – in the network layer – many functions of nowadays’ application layer and Content Distribution Networks (CDNs), like awareness of content chunks being transferred, caching, and content processing/adaptation. 

We will introduce the NDN concepts and the basic DAS over NDN mechanisms, derive upper performance bounds, and specifically discuss how a novel adaptive forwarding scheme can greatly improve performance and service quality for dynamic adaptive video streaming over NDN. Results have been gained via NDN simulations and the research group’s NDN testbed.

Brief Speaker Biography:

Hermann Hellwagner is a Full Professor of Informatics with the Institute of Information Technology, Alpen-Adria-Universität (AAU), Klagenfurt, Austria, leading the Multimedia Communication Research Group. He has edited several books, and has authored over 200 scientific papers on parallel computer architecture, parallel programming, and multimedia communication and adaptation. His research interests include distributed multimedia systems, multimedia communication, quality of service, and Information-Centric Networking (ICN). Dr. Hellwagner is a member of the IEEE, the ACM, the German Informatics Society (GI), and the Austrian Computer Society (OCG). He received many research grants from National (Austria, Germany) and European Funding Agencies and from industry. He served as a Vice President of the Austrian Science Fund (FWF) from 2013 to 2016.

Présentation de Lamine Lamali au GT Algodist le lundi 10 octobre 2016 à 14h en salle 178 du LaBRI

Bonjour à tous,

Lamine Lamali présentera un exposé au GT Algodist, le lundi 10 octobre 2016 à 14h en salle 178, intitulé : "Calcul de chemins dans les réseaux multicouches : une approche par la théorie des langages".

Résumé :
Les réseaux multicouches sont des réseaux où plusieurs protocoles coexistent sur différentes couches. Néanmoins, plusieurs architectures réseaux définissent des fonctions de conversion, encapsulation et désencapsulation de protocoles pour pallier cette hétérogénéité. Les algorithmes classiques de calcul de chemins dans les graphes ne peuvent être appliqués dans ce contexte car ils ne prennent pas en compte ces (dés)encapsulations. De plus, les plus courts chemins dans ce type de réseaux ont des caractéristiques non triviales : ils peuvent comporter des boucles et n’ont pas de sous-structure optimale.
Après avoir constaté que les séquences de protocoles associées à un chemin dans un réseau multicouche forment un langage à contexte libre, nous proposons un nouveau modèle de réseau basé sur les automates à pile, ainsi que des algorithmes issus de la théorie des langages pour calculer le(s) plus court(s) chemin(s) dans ce type de réseaux. Nous montrons également que sous contrainte de bande passante, le calcul de chemins devient NP-complet, même dans un réseau symétrique et avec seulement deux protocoles.

Tout le monde est le bienvenu.

Sémi-doc on June 22nd 2016 at 1.30pm in the LaBRI about Service Chain Modeling and Embedding for NFV-based Content Delivery

Dear all,

There will be a presentation on June 22nd, 2016 at 1.30pm (in the main amphitheater of the LaBRI) by Nicolas Herbaut on:
Service Chain Modeling and Embedding for NFV-based Content Delivery

Increasing over-the-top video consumption endangers the sustainability of content delivery over the Internet. Internet Service Providers (ISP) face difficulties in competing on value-added services with content providers and Content Delivery Network (CDN) operators. In this respect, we propose a new model for the collaboration between content delivery actors, where CDN operators can deploy their software in ISP infrastructures using Network Function Virtualization (NFV) techniques.
As the ISP network topology and utilization is deemed confidential, we use a high-level Service Level Agreement (SLA) for the negotiation of both computing resources and connectivity, allowing the ISP to optimize server selection, while providing at the same time sufficient flexibility to the CDN operators for content delivery.
Furthermore, we present a linear programming formulation and an algorithm for service chain embedding. Our simulations results show significant gains of the algorithm in terms of request acceptance rate and resource consumption. Finally, we validate the efficiency of the proposed service chain model for Virtual CDN management.
Everyone is welcome.
See you there,

Mini-GT Comet aujourd'hui à 14H30 en salle 073 / Comet mini-WG today at 2.30pm in room 073

Hi everyone,

There will be a workgroup meeting today for our Comet group at 14H30 - 2.30pm in room 073 of the Labri.

Nicolas Herbaut and Joachim Bruneau-Queyreix will present their work in progress.

The information of Nicolas' talk is given below:

Title: Content Delivery Networks as a Virtual Network Function: a win-win ISP-CDN collaboration

Abstract: The switch in video delivery from traditional medium to Over-the-Top based services has been a game changer for stakeholders of the media delivery ecosystem.  It has changed the value chain enabling Content Delivery Networks and Content Providers to take the lion's share at the expense of Internet/Network Service Providers. Even if establishing collaboration between them naturally appears as the key to provide good Quality of Service to the end-users, they struggle in finding efficient and fair ways to do so.
This paper directly tackles this problem and proposes a new model for Content Delivery Actors to collaborate over a Virtualized Infrastructure, fairly balancing the revenue stream created. We list the main challenges and the new technical opportunities to solve them, among which the deployment of a distributed Network Function Virtualization platform at the edge of the Internet Service Provider's network where a virtual Content Delivery Network is deployed.

See you soon!


Page 3 sur 41234