NAPA-WINE Final Workshop 2011

Torino, January 20-21, 2011

Schedule


 Thursday Friday
    9:00 Schedules and Overlays: Interactions Between Information Scheduling and Topology Management in P2P Streaming
Renato Lo Cigno (University of Trento)
    9:45 Binding it all together: An overview of Telco based research efforts for improved content distribution
Nicolaos Laoutaris (Telefonica Research)
    10:30 Coffee break
    11:00 Benchmarking for P2P streaming: Mission impossible or unavoidable?
Gyorgy Dan (KTH)
    11:45 P2P-based video streaming from the end-user's perspective
Phuoc Tran-Gia and Tobias Hossfeld (University of Wuerzburg)
    12:30 Lunch break
    14:00 Content Aware Searching, retrieval and Streaming in Future Internet
Emanuale Quacchio (STM Microelectronics)
14:30 Opening 14:50 Overlay-ISP cooperation for live and interactive Internet-wide distributed applications
David Griffin (UCL)
14:40 Architecture of a Network-Aware P2P-TV Application: the NAPA-WINE Approach
Emilio Leonardi (Politecnico di Torino)
15:40 P2P TV: Five years ago and today: Have we gotten any further?
Ernst Biersack (Eurecom)
15:30 Friend-to-Friend IPTV
Yong Liu (Polytechnic Institute of New York University)
16:25 Coffee Break
16:15 Coffee break 16:45 Monitoring P2P-TV traffic: Methodologies and Findings
Miklos Telek (BME) and Marco Mellia (Politecnico di Torino)
16:40 P2P-Next: Building a Decentralized Content Delivery Platform
Raul Jimenez (KTH)
17:30 Closing - DEMO
17:30 Application-Layer Traffic Optimization: how did we get here, where are we going?
Enrico Marocco (Telecom Italia)
   
18:15 First day conclusions    
20:00 Dinner    

Workshop Program

Thursday, January 20


Architecture of a Network-Aware P2P-TV Application: the NAPA-WINE Approach

Prof. Emilio Leonardi (Politecnico di Torino) - Thursday, 14:40-15:30

Download: Slides | Video (.mp4)

Abstract: TPeer to Peer streaming (P2P-TV) applications have recently emerged as cheap and efficient solutions to provide real time streaming services over the Internet. For the sake of simplicity, typical P2P-TV systems are designed and optimized following a pure layered approach, thus ignoring the effect of design choices on the underlying the transport network. This simple approach however may constitute a worry for the network providers, due to the possible congestion that P2P-TV traffic can potentially generate. In this talk we present and discuss the architecture of the innovative and network cooperative P2P-TV application that is being designed and developed within the NAPA-WINE. Our application is explicitly targeted to favor cooperation between the application and the transport network layer.

About the author:

Emilio Leonardi is currently an Associate Professor at the Dipartimento di Elettronica of Politecnico di Torino. He received a Dr.Ing degree in Electronics Engineering in 1991 and a Ph.D. in Telecommunications Engineering in 1995 both from Politecnico di Torino. In 1995, he visited the Computer Science Department of the University of California, Los Angeles (UCLA), in summer 1999 he joined the High Speed Networks Research Group, at Bell Laboratories/Lucent Technologies, Holmdel (NJ); in summer 2001, the Electrical Engineering Department of the Stanford University and finally in summer 2003, the IP Group at Sprint, Advanced Technologies Laboratories, Burlingame CA. He participated in several national and European projects such as IST-SONATA and IST-DAVID amd the NoE e-Photon-ONe, Euro-FGI. He has also been involved in several consulting and research project with private industries, including Lucent Technologies-Bell Labs., IBM, British Telecom, Alcatel and TILAB. He is the scientific coordinator of the European 7-th FP STREP project "NAPA-WINE" on P2P streaming applications, involving 11 European research institutions, operators and manufacturers. He has co-authored over 200 papers published in international journals and presented in leading international conferences, all of them in the area of telecommunication networks. He participated to the program committees of several conferences including: IEEE infocom, IEEE Globecom and IEEE ICC. He was guest editor of two special issues of IEEE Journal of Selected Areas of Communications focused on high speed switches and routers. His research interests are in the field of: Performance evaluation of Wireless Networks, P2P Streaming Systems, Queueing Theory, Packet Switching Architectures.



Friend-to-Friend IPTV

Yong Liu (Polytechnic Institute of New York University) - Thursday, 15:30-16:15

Download: Slides | Video (.mp4)

Abstract: The increasing popularity of Online Social Networks (OSN) provides new opportunities to improve P2P IPTV services. In this talk, we present two of our recent work on the integration of P2P IPTV with OSN: 1) P2P Incentive: we propose a OSN-based incentive mechanism that facilitates Networked Asynchronous Bilateral Trading (NABT) among peers to improve the efficiency of the "Tit-for-Tat" type of bilateral P2P trading. 2) Content Recommendation: we propose a Bayesian-inference based content recommendation between direct and indirect friends in OSNs.

About the author:

Yong Liu graduated with Ph.D degree from the ECE Dept. at University of Massachusetts, Amherst in 2002. He worked as a Postdoc in computer networks research group at UMass from February 2002 to February 2005. He joined ECE department of Polytechnic Institute of New York University as an assistant professor in March 2005. His current research interest includes: overlay/P2P networks, multimedia networking, network measurement, and robust network design. Yong Liu is the recipient of the Best Paper Award of IEEE INFOCOM 2009, and the 2008 Multimedia Communications Best Paper Award of IEEE Communications Society. More information about his research and teaching is available at: http://eeweb.poly.edu/faculty/yongliu/



P2P-Next: Building a Decentralized Content Delivery Platform

Raul Jimenez (KTH) - Thursday, 16:40-17:30

Download: Slides | Video (.mp4)

Abstract: P2P-Next is a large (19MEUR) FP7 integrated project whose aim is to build a content delivery platform based on peer-to-peer (P2P) technology suitable for live and on-demand video streaming. To meet these ambitious goals, the project has advanced the state of the art in this technology area. The technology developed from our research has been implemented, integrated, deployed and tested in our own testbeds as well as the open internet, both in the form of software running on PC hardware, and integrated in Pioneer developed set-top boxes. KTH's technical contributions to the project has focused on decentralized peer discovery and selection mechanims. Most recently with the aim of reducing lag when starting and switching streams, even using fully decentralized, distributed peer discovery mechanisms. Our latest implementation, currently being tested, enables our video player to start playback in less than 500ms. We are now also turning our focus towards minimizing backbone capacity use and transit costs for ISPs and content providers by considering latency, network topology and other locality parameters.

About the author:

I'm a PhD student at KTH in Stockholm since 2007. I received my computer engineering degree from the University of Zaragoza (Spain) in 2006. My main areas of interest are distributed systems, peer-to-peer and security. I am especially interested in distributed hash tables (DHTs) and topology-aware content distribution. I'm involved in the P2P-Next project. Currently, our main focus is to integrate a fast, topology-aware DHT into the BitTorrent framework. http://www.tslab.ssvl.kth.se/raul



Application-Layer Traffic Optimization: how did we get here, where are we going?

Enrico Marocco (Telecom Italia) - Thursday, 17:30-18:15

Download: Slides | Video (.mp4)

Abstract: A significant part of the Internet traffic is today generated by peer-to-peer applications such as file-sharing and live/on-demand media streaming. Such applications discover a route to each other through an overlay network with little knowledge of the underlying network topology. As a result, they choose peers randomly, or at best based on information deduced from empirical measurements that often lead to suboptimal choices. This presentation will introduce the rationale behind the Application-Layer Traffic Optimization (ALTO) working group, chartered within the IETF to define a protocol that will let ISPs provide P2P applications with topology information to perform better-than-random peer selection. It will then provide an update about the status of the working group and an introduction of the most relevant development and deployment experiences going on both in the industry and academia.

About the author:
Enrico Marocco is a Research Engineer in the Service Innovation Area at Telecom Italia. Enrico has been involved in the design and deployment of the first VoIP network at Telecom Italia, with special focus on standards compliance. He is active in IETF, where he is serving as chair of the Application-Layer Traffic Optimization (ALTO) and Call Control UUI Service for SIP (CUSS) working groups. He is also involved in various open source activities. His research interests include peer-to-peer communications and content distribution.

Friday, January 21


Schedules and Overlays: Interactions Between Information Scheduling and Topology Management in P2P Streaming

Renato Lo Cigno (University of Trento) - Friday, 9:00-9:45

Download: Slides | Video (.mp4)

Abstract: The goal of a network-aware P2P live streaming application is the distribution, upon an overlay topology, of all the information of the stream to all the peers within a certain deadline, which ensures that users have the feeling of watching a live event. The success and efficiency of the application depend on the construction of a good overlay, i.e., an overlay that guarantees that interconnected peers are close one another based on some metric while the overall cost of the system for ISPs is sustainable, as well as on the use of effective scheduling techniques for information distribution: i.e., the selection of the correct information and the correct peer to exchange it with within the peer neighborhood. Often overlay management and scheduling protocols and algorithms are not independent, as for instance in systems where information exchange is negotiated. This talk will overview some of the key results in the Napa-Wine project related to information scheduling and topology management, discussing benefits and cons of different solutions explored.

About the author:

Prof. Renato Lo Cigno is Associate Professor at the Department of Computer Science and Telecommunications (DISI) of the University of Trento, Italy, where he is one of the founding members of the Networking research group. He received a degree in Electronic Engineering with a specialization in Telecommunications from Politecnico di Torino in 1988, the same institution where he worked until 2002. From 1998 to 2001 he has been several times at the CS department of UCLA, CA, as visiting researcher for a total of more than one year. On the first visit he was supported by CNR grant No.203.15.8, and was awarded a final prize for scientific productivity. In 2001/2002 he was scientific director of INLAB, the Integrated Networking Laboratory of CSP Piemonte, now "NGN Lab" (http://www.csp.it/en/what-we-do/next-generation-networks). Renato Lo Cigno participated in several research projects funded by ESA (TOPSIM development and validation), the EU (Cost 242,247,257, Copernicus ATMIN, IST Helinet, Napoleon, and Discreet (http://ist-discreet.org)) and the Italian Ministry of University and research (MIUR) (MQOS, Planet-IP, EURO, and others). He has been the National Coordinator of the PROFILES project (http://profiles.dit.unitn.it), and he was Coordinator for Trento University in the MIUR TWELVE project (http://twelve.unitn.it). He is involved in the 7-th framework project "Napa-Wine" financed in the first call 1.5 "Networked Media" as coordinator of the Work Package devoted to the P2PTV client study and development. Renato Lo Cigno is member of the IEEE and ACM. Currently he serves in the Editorial Board of Computer Networks (Elsevier) and has been involved as Chair or TPC members of major international conferences in networking. He has co-authored over 120 papers in networking published in refereed international journals and conferences. His interest are in development, design, and performance analysis of networking systems and protocols, with particular attention to changes in communication paradigms from client/server models, to peer-to-peer models, to pervasive/ubiquitous. Since 1996 he has collaborated frequently with automatic control scientists and researchers on topics related to network control and access optimization, cooperation that lead to several publications in top international forums.



Binding it all together: An overview of Telco based research efforts for improved content distribution

Nikolaos Laoutaris (Telefonica Research) - Friday, 9:45-10:30

Download: Video (.mp4)

Preliminary abstract: In my talk I will present an overview of our research efforts in Telefonica for improving the quality of media distribution for our customers. Issues that will be covered include:
About the author:

Nikolaos Laoutaris is a senior researcher at the Internet research group of Telefonica Research in Barcelona. Prior to joining the Barcelona lab he was a postdoc fellow at Harvard University and a Marie Curie postdoc fellow at Boston University. He got his PhD in computer science from the University of Athens in 2004. His general research interests are on system, algorithmic, economic, and performance evaluation aspects of computer networks and distributed systems with emphasis on content distribution, overlay networks, P2P, and multimedia communications. For more details: http://research.tid.es/nikos/



Benchmarking for P2P streaming: Mission impossible or unavoidable?

Gyorgy Dan (KTH) - Friday, 11:00-11:45

Download: Slides | Video (.mp4)

Abstract: Benchmarking is used in many areas of computer science and computer engineering to compare the performance of algorithms and systems. Examples are image and video coding, compilers and database systems. Benchmarking requires three main factors to be available: a set of well-understood evaluation scenarios, a set of reference algorithm/system implementations, and a proper evaluation methodology. In this talk we provide examples for two of these factors and discuss how a complete benchmarking toolbox for P2P streaming systems would help future research efforts.

About the author:

György Dan received the M.Sc. degree in computer engineering from the Budapest University of Technology and Economics, Hungary in 1999 and the M.Sc. degree in business administration from the Corvinus University of Budapest, Hungary in 2003. He worked as a consultant in the field of access networks, streaming media and videoconferencing 1999-2001. He received his Ph.D. in Telecommunications in 2006 from KTH, Royal Institute of Technology, Stockholm, Sweden, where he currently works as an assistant professor. He was visiting researcher at the Swedish Institute of Computer Science in 2008. His research interests include the design and analysis of distributed and peer-to-peer systems.



P2P-based video streaming from the end-user's perspective

Prof. Phuoc Tran-Gia, Dr. Tobias Hossfeld (University of Würzburg) - Friday, 11:45-12:30

Download: Slides | Video (.mp4)

Abstract: P2P is seen as a promising technology for providing video content efficiently to the users while reducing the cost for the content providers. In this context, there are currently two trends in the Internet that are (a) locality-awareness to increase the efficiency of content distribution and (b) the diversification of end user devices which may be supported by Scalable Video Codecs (SVC). In this talk, we consider the impact of both approaches from the end-user's perspective.

(a) Locality-awareness is intended to reduce the inter-domain traffic which is costly for Internet service providers (ISPs) and simultaneously increase the performance from the viewpoint of the P2P users. We discuss simulation results studying locality-aware-implementations of BitTorrent-based video-on-demand (VoD) considering real-life, skewed peer distributions.

(b) The scalable extension of H.264/AVC offers the possibility to adapt the user perceived video quality on-the-fly to the network situation and access capabilities of the user devices. First, we propose and evaluate a P2P VoD architecture based on the Tribler application which is enhanced to support such a SVC video. Then, we discuss the impact of uncontrollable influence factors such as packet loss and controllable influence factors such as a lower frame rate or a lower resolution on the Quality of Experience (QoE) as perceived by the end user.

About the authors:
Prof. Phuoc Tran-Gia is Chair of Communication Networks within the Institute of Computer Science, University of Würzburg, Germany. His current research areas include architecture and performance analysis of communication systems, and planning and optimization of communication networks. He has been active in several management committees of European research projects. He is currently working with the European Union authorities and the German Ministry of Research and Education on funding strategies and initiatives towards Next Generation Internet. He is coordinator of the project German-Lab (G-Lab), aiming to foster experimentally driven research to exploit future internet technologies.
Dr. Tobias Hossfeld studied computer science and mathematics at the University of Würzburg, Germany. He finished his PhD on performance evaluation of future Internet applications and emerging user behavior in 2009. Currently, he is heading the FIA research group "Future Internet Applications & Overlays" at the Chair of Communication Networks in Würzburg. His main research interests cover network virtualization, social networks, self-organization mechanisms in overlay networks and P2P systems, with a special focus on mobile environments like 3G and 4G, as well as investigations on Quality of Experience (QoE) for Internet applications like Skype, YouTube, or Web Browsing in general.



Content Aware Searching, retrieval and Streaming in Future Internet

Emanuele Quacchio (STMicroelectronics) - Friday, 14:00-14:50

Download: Slides | Video (.mp4)

Abstract: The COAST project aims to build a Future Content-Centric Network (FCN) overlay architecture able to intelligently and efficiently link billions of content sources to billions of content consumers, and to offer fast content-aware search, retrieval, delivery and streaming. The presentation will describe in details key enablers of the COAST project, main technologies deployed, progresses achieved and the status of the implementation activities.

About the author:

Emanuele Quacchio received the Master Degree in Electronic Engineering from Polytechnic University of Turin, Italy in 2003. He worked two years as a researcher in the Dept. of Electronics of the same university and joined STMicroelectronics at the AST-system R&D group in 2006, where he is currently working as a Senior System Engineer. His activities are mainly focused on embedded SW development for STB/mobile platforms, video compression standards and streaming protocols. He published and co-authored several papers on the principal journals of engineering and conferences. Since 2006 he has participated to a number of EU funded projects (ASTRALS, SEA, P2PNext, COAST).



Overlay-ISP cooperation for live and interactive Internet-wide distributed applications

David Griffin (UCL) - Friday, 14:50-15:40

Download: Slides | Video (.mp4)

Abstract: Future networked media applications will be multi-sourced, highly interactive distributed meshes of HD and 3D multi-sensory channels. While high-quality user-centric applications such as these offer tremendous advantages to their participants and to society at large, they present a major headache to ISPs as they demand unprecedented quantities of network resources in unpredictable locations. The traditional solution of pre-provisioning sufficient network resources everywhere is no longer economically viable given the huge capacities required. If these services are ever to become a reality then a fundamentally different approach is required. This talk will consist of two parts. The first is on the ongoing research activities in the ENVISION project, the second is on the research challenges in Future Media Networks as seen by the FMN-cluster of projects in the Networked Media Systems unit.

About the author:

David Griffin is a Principal Research Associate in the Department of Electronic and Electrical Engineering, University College London. He has a BSc from Loughborough University and a PhD from UCL, both in Electrical Engineering. He has had a significant involvement in EU research programmes in RACE, ACTS, IST and ICT projects with both technical research contributions and project management activities. His research interests are in the planning, management and dynamic control for providing QoS in multiservice networks, p2p networking and novel routing paradigms for the Future Internet. He is the project manager of the ENVISION project and coordinator of the Future Media Networks cluster.
More information at: http://www.ee.ucl.ac.uk/~dgriffin/.



P2P TV: Five years ago and today: Have we gotten any further?

Ernst Biersack (EURECOM) - Friday, 15:40-16:25

Download: Video (.mp4)

Abstract: P2P TV has been announced as one of the new killer applications for P2P. Five years ago there have been some promising first experiments such as PPLive and annoucements of systems such as Joost that have never fully materialized. If we look at the situation today, we realize that P2P TV has not really taken off. On the other hand, a number of other Internet-based TV distribution techiques are in use today and offer a large choice of channels. We will discuss the technical requirements for P2P TV and alternative TV distribution techniques and ask if P2P is really a viable approach for TV distribution.

About the author:

Ernst Biersack studied Computer Science at the Technische Universität München and at the University of North Carolina at Chapel Hill. He received his Dipl. Inform. (M.S.) and Dr. rer. nat. (Ph.D.) degrees in Computer Science from the Technische Universität München, Munich, Germany, and his Habilitation a Diriger des Recherches from the University of Nice, France. From March 1989 to February 1992 he was a Member of Technical Staff with the Computer Communications Research Group of Bell Communications Research, Morristown, US. Since March 1992 he has been a Professor in Telecommunications at Eurecom, in Sophia Antipolis, France. His current Research is on Peer-to-Peer Systems and Network Tomography.
His home page can be found at: http://www.eurecom.fr/~btroup/erbi_page.html



Monitoring P2P-TV traffic: Methodologies and Findings

Miklos Telek (BME), Marco Mellia (Politecnico di Torino) - Friday, 16:45-17:30

Download: Slides

Abstract: In the Internet, P2P-TV applications are readily available and users can freely watch live streams using such technology. It is then interesting for both the research community and the ISP to understand the impact that P2P-TV traffic can have on the network, e.g., by characterizing and modeling such complex systems. Unfortunately, all successful applications exploit proprietary protocols and unknown algorithms, so that it is even difficult to quantify the popularity of different P2P-TV systems. In this talk we will present first a set of instruments to investigate the behavior of P2P-TV applications, such as traffic classification techniques and data collection tools. Then we will present an extensive characterization of traffic P2P-TV applications inject in the network.

About the authors:

M. Telek received the M. Sc. degree in Electrical Engineering from theTechnical University of Budapest in 1987. In the same year he joined the Hungarian Post Research Institute where he studied the modelling, analysis and planning aspects of communication networks. Since 1990 he has been with the Department of Telecommunications of the Technical University of Budapest, where he is a full professor now. He received the C.Sc. and D.Sc. degree from the Hungarian Academy of Science in 1995 and 2004, respectively. His current research interest includes stochastic performance modeling and analysis of computer and communication systems.

Marco Mellia received the Ph.D. degree in telecommunications engineering from Politecnico di Torino, Torino, Italy, in 2001. In 1999, he was with the Computer Science Department, Carnegie Mellon University, Pittsburgh, PA. Since April 2001, he has been with the Electrical Engineering Department, Politecnico di Torino. He has coauthored over 150 papers published in international journals and conferences, and he participated in the program committees of several conferences including IEEE INFOCOM and ACM SIGCOMM. He is the Work Package coordinator of the Napa-Wine project. His research interests are in the fields of traffic measurement, P2P applications, and energy-aware network design.




Page created by Armin Jahanpanah - Last update: Feb 7, 2011.