Program for ACM Workshop on Advanced Video Streaming Techniques for Peer-to-Peer Networks and Social Networking

Workshop website: http://www.p2pstreaming.eu/
Workshop date/location: 29th October 2010, Florence, Italy (co-located with ACMMM'10)
ACM Multimedia: http://www.acmmm10.org/

Workshop presentations: No central server will be used for oral presentations at the Conference Workshops. Authors can either use their own laptops or the PC available in each Workshop room (Windows based machine). Authors are invited to check this program page to see the time slots allocated for oral presentations. For poster and demo sessions at the Workshops, boards are available that can hold a poster up to 100 cm width x 250 cm length (3.28 ft x 8.2 ft). 

Finally, we also invite you to refer to the Conference web page (http://www.acmmm10.org/) for any further information about venue, travel, weather, accommodations, restaurants, and so on.

Final Program:

09:00-09:15: Welcome Address by the Chairs
09:15-10:00: Keynote Address
Session Chair: Christian Timmerer (Klagenfurt University, Austria)
Audio/visual content and metadata delivered over the open Internet using P2P-Next: some experiences from a broadcaster's perspective by George Wright, Head of Prototyping, BBC Research and Development

10:00-10:30: Short presentation of posters and demos

10:30-11:00: Coffee break

11:00-12:00: Session 1 - Networking and Streaming
Session Chair: Christian Timmerer (Klagenfurt University, Austria)

• P2P Group Communication with Layer-Aware FEC by Yago Sánchez; Cornelius Hellge; Thomas Schierl; Thomas Wiegand (Fraunhofer Institute for Telecommunications - Heinrich-Hertz-Institute, Germany)
• P2P streaming with LT codes: a prototype experimentation by Andrea Magnetto; Rossano Gaeta; Marco Grangetto; Matteo Sereno (Universita' di Torino, Italy)
• Peer-to-Peer streaming based on network coding improves packet jitter by Riccardo Bernardini; Roberto Cesco Fabbro; Roberto Rinaldo (University of Udine, Italy)

12:00-13:00: Session 2 - Application
Session Chair: Pascal Frossard (EPFL, Switzerland)

• A Novel Cache Optimization Algorithm and Protocol for Video Streaming in Pure Peer-to-Peer Networks by Carlo Giulietti; Dan Schonfeld; Rashid Ansari (University of Illinois at Chicago, USA)
• Access Control to BitTorrent swarms using Closed Swarms by Keith Mitchell (Lancaster University, United Kingdom); Njaal Borch (Norut IT, Norway); Ingar Arntzen (Norut IT, Norway); Dusan Gabrijelcic (Jozef Stefan Institute, Slovenia)
• Advanced Prefetching and Upload Strategies for P2P Video-on-Demand by Osama Abboud; Konstantin Pussep; Markus Müller; Aleksandra Kovacevic; Ralf Steinmetz (Technische Universitaet Darmstadt, Germany)

13:00-14:30: Lunch

14:30-15:30: Session 3 - Design
Session Chair: Keith Mitchell (University of Lancaster, UK)

• A Hybrid Approach to Modeling End-to-End Delay in P2P Networks by Philipp Berndt; Dominic Battré; Odej Kao (Technische Universität Berlin, Germany)
• Design and Implementation of a Generic Library for P2P Streaming by Luca Abeni; Csaba Kiraly; Alessandro Russo; Marco Biazzini; Renato Lo Cigno (University of Trento, Italy)
• Design and Evaluation of an Optimized Overlay Topology for a Single Operator Video Streaming Service by Stefano Giordano (University of Pisa, Italy); Rosario G. Garroppo (University of Pisa, Italy); Stella Spagna (University of Pisa, Italy); Saverio Niccolini (NEC Europe Ltd., Germany); Jan Seedorf (NEC Europe Ltd., Germany)

15:30-16:00: Coffee break

16:00-18:00: Session 4 - Posters and Demos
Session Chair: Gabriella Olmo (Politecnico di Torino, Italy)
Posters:

• An Analytical Approach to Model Adaptive Video Streaming and Delivery by Razib Iqbal; Shervin Shirmohammadi (University of Ottawa, Canada)
• Consistent Image Decoding from Multiple Lossy Versions by Marco Dalai; Serena Malavasi; Riccardo Leonardi (University of Brescia, Italy)
• Improving Quality-of-Experience for Multiple Description Video Transmission in Peer-To-Peer Networks by Simone Milani; Giancarlo Calvagno (University of Padova, Italy)
• Knapsack Problem-based Piece-Picking Algorithms for Layered Content in Peer-to-Peer Networks by Michael Eberhard (Klagenfurt University, Austria); Tibor Szkaliczki (Computer and Automation Research Institute, Hungarian Academy of Sciences, Hungary); Hermann Hellwagner (Klagenfurt University, Austria); Laszlo Szobonya (Computer and Automation Research Institute, Hungarian Academy of Sciences, Hungary); Christian Timmerer (Klagenfurt University, Austria)
• MixNStream: Multi-Source Video Distribution with Stream Mixers by Philip Chun Ho Yuen; Gary Chan (The Hong Kong University of Science and Technology, P.R. China)
• Three Highly Available Data Streaming Techniques and Their Tradeoffs by Sumita Barahmand; Shahram Ghandeharizadeh; Anurag Ojha; Jason Yap (USC, USA)

Demos: (tentative)

Closed Swarms by Njaal Borch (Norut IT, Norway)

Abstract: In the search for commercial opportunities within online media distribution, Closed Swarms has been designed to provide content providers with a flexible, distributed authentication mechanism for P2P media distribution. Closed Swarms allow the provider control of bandwidth costs, yet it can enable new business models which might be better suited for the Internet Age, such as freemium solutions or added benefits for paying customers. We will demonstrate how Closed Swarms can be used to provide a free, non-guaranteed community service while logged-in users are provided with a full VOD experience.

Visualizing and Reducing Wait Delay in Periodic Butterfly Communication by Philipp Berndt (Technische Universität Berlin, Germany)

Abstract: Butterfly graphs are probably best known from their use in the Fast Fourier Transform (FFT) algorithm. A different use is in P2P VoIP Conferencing, where each node combines the audio streams it receives, before forwarding the data to the next stage. Dead time between the reception and the relaying of audio data adds up along the data paths to an overall latency that impairs the communication experience. This wait delay heavily depends on the send phase offsets between the nodes. Our demonstration shows how this problem can be interactively visualized and studied, what relationships exist and which methods can be employed to minimize overall wait delay.

Developing P2P Streaming Applications with GRAPES by Luca Abeni (University of Trento, Italy)

Abstract: In this demonstration, it will be shown how to use the GRAPES library to build a simple P2P application, either single-threaded or multi-threaded (the demo application will show how the GRAPES API allows to change the structure of the application from single-threaded to multi-threaded by simply modifying the main application loop). It will also be shown how to easily change the implementation of a GRAPES module (the peer sampler), and the impact of such a change on some performance metrics will be evaluated.

NextShareTV demo by P2P-Next (http://www.p2p-next.eu)

Abstract: The NextShareTV is a Set-top-box confirming to the NextShare platform, developed as part of the P2P-Next project. As a NextShare device, it collaborates with NextSharePC implementations as well, bridging the TV set and the PC in a single content swarm.

18:00-18:30: Best Paper Aware + Closing

Multimedia conferences this week

I just saw on my calendar that this week there are a couple of multimedia-related events worth to report here:

• MMedia2010 (June 13-19, 2010 - Athens/Glyfada, Greece), 2nd International Conference on Advances in Multimedia, where our department had a couple of papers in the program ranging from self-organizing multimedia systems and transcoding to peer-to-peer architectures.
• WoWMoM2010 (June 14-17, 2010 - Montreal, Canada), International Symposium on a World of Wireless, Mobile and Multimedia Networks, where the program covers all layers in the network related to wireless and multimedia. Interestingly, the program includes a PhD forum and industry track.
• IWQoS2010 (June 16-19, 2010 - Beijing, China), International Workshop on Quality of Service, where Henning Schulzrinne is giving a keynote entitled "25 years of quality of service research - where next?". In general, this workshop is the QoS workshop in the world and the program is available here (there's a session about overlays and peer-to-peer networks).
• FMN2010 (June 17-18, 2010 - Krakow, Poland), 3rd International Workshop on Future Multimedia Networking, where the program is only available as PDF and there's also a session on QoS and Quality of Experience (QoE).

ACM Workshop on Advanced Video Streaming Techniques for Peer-to-Peer Networks and Social Networking

Camere-ready deadline extension: July 26, 2010
Workshop held within ACM Multimedia, 25-29 October 2010, Firenze, Italy

http://www.p2pstreaming.eu
[PDF | TXT]

Peer-to-peer (P2P) is a promising technology for video streaming, and offers advantages in terms of robustness, re-configurability and scalability.  In this context, social networks and social services are emerging as a potential new driver for content delivery networks. Specifically, social networks potentially provide a new level of understanding and knowledge related to the interaction between people within a virtual space. Many emerging multimedia based services and applications have started to exploit the ‘social graph’ in new ways for establishing a basis for social recommendations, filtering etc.). As yet, one unexplored area of research relates to the exploiting the social graph for informing adaptive behaviour in P2P-based multimedia systems. On the other hand, the P2P video technology is still challenging, due to the need of reducing start-time and churn-induced instability, to the asymmetry of residential broadband connections, and to high packet loss rates due to router congestion and transmission errors on the physical network, node departure from the P2P overlay, strict timing out due to real time visualization. The lack of guarantee about the actual delivery of the data may cause drops in the reproduction quality and service outages. The workshop objective is to solicit novel contributions on all aspects of P2P-based video coding, streaming, and content distribution which is
informed by social networks.

Topics of interest include, but are not limited to

• Innovative P2P-based video streaming solutions
• P2P-based social media content distribution networks
• Advanced video coding techniques for real-time P2P applications: layered/scalable video coding, multiple description coding, distributed source coding
• Identification and design of proper metrics for performance evaluation and monitoring including Quality of Service/Experience
• Content and context analysis and modelling for P2P-based social media distribution
• Filtering and recommendation systems
• Error-resilience tools for peer-to-peer multimedia services
• Rate control and bandwidth adaptation for both single streams and multiple stream multiplexing
• Cross layer optimization issues
• Protocols for peer-to-peer multimedia services
• P2P streaming prototype implementation for both live and on-demand streaming
• Advertisement, payment, and cashing systems
• Applications, standards, and practical deployments

Invited Talk

Audio/visual content and metadata delivered over the open Internet using P2P-Next: some experiences from a broadcaster's perspective

by George Wright, Head of Prototyping, BBC Research and Development

The best paper award (€300) is sponsored by RADVISION (http://www.radvision.com/).

Important Dates

• Paper submission: June 1014, 2010 (extended)
• Notification of acceptance: July 10, 2010
• Camera-ready submission: July 2026, 2010 (due to constraints imposed by ACM)
• Workshop date: October 29, 2010

Submission Guidelines
Papers should not exceed five pages in length, following the ACM proceedings format. Papers must be original and have not been published or under consideration for publication elsewhere. Each paper will be reviewed by three members of the program committee, recognised for their competence in the field. Final decision about inclusion in the workshop proceedings will be taken by the PC members exclusively on the basis of the obtained reviews and levels of recommendation. Papers must be registered using the ACM Multimedia conference management software (http://www.edas.info/N9045).

Accepted papers will be included in the workshop's proceedings which will be published together with the proceedings of the ACM Multimedia Conference 2010. Additionally, a special issue or an edited volume is planed based on the best papers of the workshop.

Workshop Organisers:
  Gabriella Olmo, Politecnico di Torino
  Christian Timmerer, Klagenfurt University
  Pascal Frossard, EPFL
  Keith Mitchell, University of Lancaster

Demonstrations
  Njål Borch, NORUT, Norway

Program Committee
  Jari Ahola, VTT, Finland
  Peter Amon, Siemens, Germany
  Riccardo Bernardini, Università di Udine, Italy
  Njål Borch, NORUT, Norway
  Giancarlo Calvagno, Università di Padova, Italy
  Luca Celetto, STMicroelectronics, Italy
  Pablo Cesar, CWI, The Netherlands
  Jaime Delgado, Universitat Politècnica de Catalunya, Spain
  Marek Domanski, Poznan University of Technology, Poland
  Kalman Graffi, TU Darmstadt, Germany
  Marco Grangetto, Università di Torino, Italy
  Hermann Hellwagner, Klagenfurt University, Austria
  Ralf Klamma, RWTH Aachen, Germany
  Mathias Lux, Klagenfurt University, Austria
  Enrico Magli, Politecnico di Torino, Italy
  Pierangelo Migliorati, Università di Brescia, Italy
  Daniel Negru, University of Bordeaux, France
  Jens-Rainer Ohm, RWTH Aachen, Germany
  Jörn Ostermann, Leibniz Universität Hannover, Germany
  Giovanni Pau, University of California at Los Angeles, US
  Fernando Pereira, IST, Portugal
  Roberto Rinaldo, Università di Udine, Italy
  Thomas Schierl, HHI, Germany
  Iraj Sodagar, Microsoft, US
  Marco Tagliasacchi, Politecnico di Milano, Italy
  Tammam Tillo, Xi’an Jiaotong-Liverpool University, China
  Stefano Tubaro, Politecnico di Milano, Italy
  Theodore Zahariadis, Synelixis Solutions Ltd, Greece

The workshop is partly sponsored by the project “ARACHNE: Advanced video streaming techniques for peer-to-peer networks,” funded by the Italian Ministry for Education and Research (www.diegm.uniud.it/arachne), and partly by the EC-funded “P2P-Next” project (www.p2p-next.org).

11th Multimedia Metadata Community Workshop on Interoperable Social Multimedia Applications (WISMA 2010)

On May 19-20, 2010 I've attended the 11th Multimedia Metadata Community Workshop on Interoperable Social Multimedia Applications (WISMA 2010) in Barcelona, Spain. The proceedings are available online on CEUR-WS.org and Twitter stream in case you'd like to review it.

I had two presentation which I'd like to provide here. The first one was on A Metadata Model for Peer-to-Peer Media Distribution:

Abstract: In this paper we describe a metadata solution for a Peer-to-Peer (P2P) content distribution system termed NextShare. We outline the key motivating factors for our approach, detail the overall generic architecture we have developed and present the workflow for delivering metadata through Peer-to-Peer based content distribution. The paper also presents the metadata model we have developed and we describe in detail how all the content can be packetized and distributed using NextShare. Finally, a description of the core and optional metadata attributes which may be utilized within the system is provided.

A Metadata Model for Peer-to-Peer Media Distribution

View more presentations from Klagenfurt University.

The second presentation provided an answer for the following question: Are Sensory Effects ready for the World Wide Web?

Abstract. The World Wide Web (WWW) is one of the main entry points to access and consume Internet content in various forms. In particular, the Web browser is used to access different types of media (i.e., text, image, audio, and video) and on some platforms is the only way to access the vast amount of information on the Web. Recently, it has been proposed to stimulate also other senses than vision or audition while consuming multimedia content through so-called sensory effects, with the aim to increase the user’s Quality of Experience (QoE). The effects are represented as Sensory Effects Metadata (SEM) which is associated to traditional multimedia content and is rendered (synchronized with the media) on sensory devices like fans, vibration chairs, lamps, etc. In this paper we provide a principal investigation of whether the sensory effects are ready for the WWW and, in anticipation of the result, we propose how to embed sensory effect metadata within Web content and the synchronized rendering thereof.

Are Sensory Effects ready for the World Wide Web?

View more presentations from Klagenfurt University.

Peer-to-Peer Computing 2010

August 25-27, 2010, Delft, Netherlands: http://p2p10.org/

Scope and Topics
The P2P'10 conference solicits papers on all aspects of peer-to-peer computing. Of particular interest is research that furthers the state-of-the-art in the design and analysis of peer-to-peer applications and systems, or that investigates real, deployed, large-scale peer-to-peer applications or systems. The applications can range from the traditional peer-to-peer application of file sharing to more novel applications such as media streaming and intelligence sharing. Topics of interest include but are not limited to the following aspects of P2P computing:

• Information retrieval and query support
• Semantic overlay networks and semantic query routing
• P2P for grids, clouds, and datacenters
• Deployed (commercial) applications and systems
• Security, trust, and reputation
• Cooperation, incentives, and fairness
• P2P economics
• Social networks
• Overlay architectures and topologies
• Overlay interaction with underlying infrastructure
• Overlay monitoring and management
• Self-organization
• P2P applications and systems over mobile networks
• Performance, robustness, and scalability 

Paper submission guidelines
Papers can be submitted either as full papers or as short papers in the IEEE single-spaced two-column format using at least 10 point font size. Full papers should not exceed 10 pages and short papers should not exceed 4 pages. Papers must be submitted electronically in PDF format through the paper-submission website linked from the conference website. IEEE templates for LaTeX and Microsoft Word, as well as related information, can be found at the IEEE Digital Toolbox webpage. The conference proceedings will be published by the IEEE Communications Society.

All submissions will be evaluated using a double-blind review process. To ensure blind reviewing, papers should be anonymized by removing author names and affiliations, as well as by masking any information about projects and bibliographic references, etc. that might reveal the authors' identities. Papers that are not properly anonymized will be returned without review. Submitted papers should describe original and previously unpublished research and are not allowed to be simultaneously submitted or under review elsewhere.

All papers must be registered with the submission system by the abstract submission deadline. At least one author of every accepted paper must register to the conference not later than June 10, 2010, 12 PM (EST) and present the paper.

Important Dates

• Abstracts Due March 29, 2010
• Full Papers Due April 5, 2010
• Conference: August 25, 2010

Peer-to-peer for content delivery for IPTV services: analysis of mechanisms and NGN impacts

Just found this document here which is a (first) draft of ETSI TISPAN's work in the area of P2P television. It seems that P2P is becoming a major issue for standardization:

• IETF has mainly ALTO and P2PSIP working groups but also P2P streaming is pushing "on the market."
• DVB recently released an Internet TV questionnaire which has some P2P aspects.
  
• W3C had some P2P activity in the past but now it seems to be silent with respect to this topic. Maybe the situation changes in the future...
  
• MPEG provides the baseline technologies (codecs and file formats) but also started an exploration activity towards an Advanced IPTV Terminal (AIT), probably jointly with ITU-T.

The future will show us which of these activites will be successful and I'd like to close with a quote that fits here very well, i.e., "The nice thing about standards is that you have so many to choose from". --Andrew S. Tanenbaum

Let me know in case something is missing or wrong.

Problem Statement of P2P Streaming Protocol (PPSP)

"We propose to develop an open peer-to-peer (P2P) streaming protocol named PPSP. This document describes the problems related to PPSP and outlines considerations that have to be taken in account when arriving at equitable solutions." ... this is the abstract but what's the actual scope of the problem statement?

• Proprietary protocols and an open PPSP
• Integrating cache and CDN into P2P streaming
• Integrating existing protocols into P2P streaming
• Mobility and wireless issue such as end to end communication is harder and limited bandwidth resource

The design issues include the choice of architecture and integration with existing protocols (i.e., legacy issues) such as RELOAD, ALTO, RTSP (and encapsulation), and edge devices (e.g., cache integration).

Application-Layer Traffic Optimization (ALTO) Problem Statement

Peer-to-peer applications, such as file sharing, real-time communication, and live media streaming, use a significant amount of Internet resources. Such applications often transfer large amounts of data in direct peer-to-peer connections. However, they usually have little knowledge of the underlying network topology. As a result, they may choose their peers based on measurements and statistics that, in many situations, may lead to suboptimal choices. This document describes problems related to optimizing traffic generated by peer-to-peer applications and associated issues such optimizations raise in the use of network-layer information.

A URL for this Internet-Draft is:
http://www.ietf.org/internet-drafts/draft-ietf-alto-problem-statement-01.txt

Internet-Drafts are also available by anonymous FTP at:
ftp://ftp.ietf.org/internet-drafts/

Analytical Model for BitTorrent- based Live Video Streaming

Just found this article which is worth reading especially if you're working in the area of P2P live video streaming.

Tewari, S. and Kleinrock, L., "Analytical Model for BitTorrent- based Live Video Streaming", in Proceedings of IEEE NIME 2007 Workshop, Las Vegas, NV, Jan 2007. PDF

Abstract: Peer-to-peer live video streaming over the Internet has been measured to support over 100,000 concurrent users. While the approach is very attractive, established providers need to understand the performance of such a system before deploying such a system as a frequent loss in quality would jeopardize their reputation. This paper provides an analytical model to inform the design of BitTorrent-based live video streaming solutions. While, given the current broadband deployment scenario, a pure peer-to-peer solution can support only limited streaming rates, our analysis shows that the addition of a well-designed peer-to-peer solution to existing server-based streaming infrastructures can allow substantially higher streaming rates. The efficiency of a BitTorrent-like peer-to-peer solution depends on the peer group size and the number of fragments available for sharing at any given time. Our analysis suggests that the efficiency of the peer-to-peer solution is not sensitive to the size of the peer group for groups larger than 15-20 users. A similar threshold exists for the number of fragments available for sharing at any given time. For live streaming scenarios, this threshold dictates that the fragment size be substantially smaller than the default fragment size in BitTorrent to ensure that the stream latency is small.

Peer-to-Peer (P2P): Is there really a 'war' between ISPs and application providers?

There's an interesting discussion ongoing at IETFs' ALTO mailing list questioning whether there's really a war between the ISPs and application (i.e., P2P) providers. The reason for this discussion is a presentation at the IETF meeting and the relevant slides can be found here. The background is as follows: ISPs do not disclose topology information - there're many possible reasons why - which would be beneficial for P2P systems. On the other hand, P2P systems do not disclose information about their users - mainly for privacy reasons - which would ease the management of the P2P traffic at the level of the ISP:

"...while the ISPs have constraints about what they can disclose, the ISPs are still able to provide useful policy information to the P2P applications. P2P applications have concerns about the privacy of their users, but do want to cooperate with the ISP to optimize network traffic in a manner beneficial to ISPs." --from here.

Currently, there are two solutions to this problem available. (1) the oracle service developed by Deutsche Telekom Laboratories and (2) P4P developed at Yale University. The authors (Comcast/BitTorrent) of this email state that they have another solution or is that based on oracle service or P4P? However, I think that IETF is the right place to discuss this and I'm looking forward to an interesting discussion which will hopefully lead to a useful RFC.

P2P Streaming Protocols at IETF v2

I recently blogged about P2P Streaming Protocols at IETF. Now the slides of this bar BoF meeting are available here. The minutes one can find on the list archives for those who are not subscribed. It is difficult to draw real conclusions at this point in time as much things are not yet clear enough. From the meeting minutes one can extract "still have uncertainties of the goal of PPSP", "PPSP should "narrow" its scope", and "proceed with a proper BOF at the next IETF". Thus, it is suggested to work "on an architecture document covering different components and analyze each components wrt to the work done or current being work upon by other WG in IETF".

P2P Streaming Protocols at IETF

It seems that the IETF is starting to form a P2P Streaming Protocols activity as there will be a BoF meeting during 74th IETF in San Francisco in March 2009. The agenda covers the following items:

• Problem Statement
• Protocol Requirements
• Protocol Analysis of PPlive and PPStream by Interent Measurement
  
• Introduction of Distributed Services Network, a vision of making PPSP and P2PSIP into reality

Furthermore, it seems that China Mobile and Huawei Technologies are the driving forces behind this exciting activity. There's also an email reflector to which one can subscribe here. I've just subscribed an I'm eager to learn more about what's going on there...

P2P News from IETF

As some people know the IETF is taking some actions w.r.t. the standardization of protocols related to Peer-to-Peer (P2P) technology within ALTO and P2PSIP groups. Recently, two new drafts have been submitted which might be interesting to some of you:

The first one is entitled An Architecture of ALTO for P2P Applications with the following abstract: ALTO enables Internet Service Providers (ISPs) and network application software distributors to work jointly and cooperatively to reduce network resource consumption and to improve application performance. In this document, we specify an architecture for integrating ALTO into peer-to-peer (P2P) applications.

The second one is related to P4P Protocol Specification: Provider Portal for Network Applications (P4P) is a framework that enables Internet Service Providers (ISPs) and network application software developers to work jointly and cooperatively to optimize application communications. The goals of this cooperation are to reduce network resource consumption and to accelerate applications. To achieve these goals, P4P allows ISPs to provide network information and guidance to network applications, allowing clients to exchange data more effectively. This document specifies the P4P protocol operations and message formats. The goal is provide a formal specification for developers to create inter-operable implementations.

Please note that latter is mainly driven by Yale University and is in competition with the so-called Oracle Service developed by an institute of the Deutsche Telekom Laboratories, namely the research group of Anja Feldmann. I wonder whether they will start/join the "standardization war".

Peer-to-peer (P2P) Architectures

Peer-to-peer (P2P) Architectures, draft-iab-p2p-archs-00.txt

Abstract (copy/paste from the draft):

In this document we provide a survey of P2P (Peer-to-Peer) systems. The survey includes a definition and a taxonomy of P2P systems. This survey also includes a description of which types of applications can be built with P2P technologies and examples of P2P applications that are currently in use on the Internet. Finally, we discuss architectural tradeoffs and provide guidelines for deciding whether or not a P2P architecture would be suitable to meet the requirements of a given application.

Very interesting survey with a lot of useful pointers, among them I'd like to mention Peltotalo, J., Harju, J., Jantunen, A., Saukko, M., and L. Vaeaetaemoeinen, "Peer-to-Peer Streaming Technology Survey", Seventh International Conference on Networking, Cancun, Mexico, pp. 342-350 , April 2008. At the end one can find a summary of existing peer-to-peer streaming technologies.

The P2P architectures' draft also provides a definition of a P2P system. However, this definition encompasses mainly that

"peers share their processing and storage capacity (i.e., their hardware and software resources) so that the system can provide a service"

In my view this is too weak and I'd like to provide a quote which I've reported some time ago that defines a P2P system as follows:

"(1) peers act as client and server, (2) they provide computational/storage resources to other peers, and (3) they're self-organizing and scaling."

The third property is the most important one as (1) and (2) is also true for several proxies out there...

IEICE Transactions on Communications: Special Section on Peer to Peer Networking Technology

The IEICE Transactions on Communications issued a special section on Peer to Peer networking technology with the following papers:

• A Remedy for Network Operators against Increasing P2P Traffic: Enabling Packet Cache for P2P Applications. In this paper the authors propose a network layer packet-level caching for reducing the volume of emerging P2P traffic but transparently to the P2P applications. However, doesn't this need a tight cooperation with the network provider (or ISP) which isn't that easy in case of P2P traffic?
  
• Incentive Mechanism for P2P Content Sharing over Heterogenous Access Network. An incentive mechanism called EMOTIVER assign rewards to users according to their access network used.
  
• CAT: A Cost-Aware BitTorrent. In this paper the authors propose to utilize inexpensive inter-ISP connections in order to reduce the networking costs but, again, this requires interaction with the ISP...
  
• An Analytical and Experimental Study of Super-Seeding in BitTorrent-Like P2P Networks. In this paper, the authors present an analytical and experimental study over the performance of super-seeding scheme with the aim to provide answers to the following questions: whether and how much super-seeding saves uploading cost, whether the overall downloading time is decreased by super-seeding, and in which circumstances super-seeding performs worse. The experiments have been performed on 250 PlanetLab nodes.
• Improving Success Ratio of Object Search in Highly-Dynamic Mobile P2P Networks. Improves the usage of P2P technology in Mobile Ad Hoc Networks (MANETs) to help finding the right objects. Evaluation by simulation...
  
• Peer-to-Peer Based Fast File Dissemination in UMTS Network. Sounds interesting too but also "only" simulations. I wonder how this works in a real world testbed...
  
• Cheatproof Dual-Tree Application-Level Multicast for Bulk Data Distribution. Also, in this paper simulation results show that their protocol outperforms the existing protocol from the view point of throughput and resource utilization against member cheating.

An interesting issue but unfortunately no paper addresses issues related to streaming media in P2P networks which, I think, is one of the most exciting topics right now...

P2P-Next Project presents at IBC 2008 World's First video end-to-end streaming distribution of professional content to low-cost STB hardware

-- see also P2P-Next news

Amsterdam, 11 September 2008 - P2P-Next, a pan-European conglomerate of 21 industrial partners, media content providers and research institutions , is releasing its first version of NextShare, an Open Source P2P (Peer-to-Peer) video delivery platform today at IBC 2008. NextShareTV is a STB (Set-Top-Box) implementation of the NextShare content delivery platform running on a low-cost embedded hardware.

The P2P-Next research project has successfully released and tested the first Beta version of their P2P live streaming technology. This new technology allows everyone to broadcast a live stream, such as a Web cam feed or TV channel, to millions of Internet users. Key is the bandwidth efficiency of this technology, by expanding the proven BitTorrent protocol you can stream to thousands of people using roughly the same amount of bandwidth as for a single user..

NextShareTV is demonstrated at the IBC 2008 exhibition on the EBU Village (Hall 10 Both D21).

”This is an extremely ambitious project which will create a viable broadcast platform that would enable large audiences to stream and interact with live and on-demand (VoD) content via a set top box or a TV receiver. In addition, it is our intention to allow audiences to build communities around their favourite content via a fully personalized system. We are enthusiastic about the P2P-Next plans to test the P2P-Next system for major international broadcasting events, “ declares Lieven Vermaele, EBU Technical Director.

"The harnessing of Open Source P2P video streaming technology like NextShare by low-cost consumer electronics represents a sea change and an exciting opportunity for the market. Pioneer recognises that consumers are demanding greater choice and quality content on demand, together with integrated devices that are easy to use. The next step is to agree Open Standards for interoperability between CE peer devices across Europe.", says Mark Stuart of Pioneer Digital Design and Technical Director of the P2P-Next project.

George Wright of BBC, Creative Director of the P2P-Next project states: “The BBC is excited to be part of a world's first end-to-end streaming of live TV via a P2P network to a set top box using professional content at professional quality. Working with partners around Europe and content producers from the Public Service Broadcasters of European Broadcasting Union (EBU), commercial and microformat areas will allow a new distribution system to be tested, evaluated and assessed.”

The P2P-Next project will run over four years, and plans to conduct large-scale technical trials of new media applications running on a wide range of consumer devices

The project has an open approach towards sharing results. All core software technology will be available as open source, enabling new business models. P2P-Next will also address a number of outstanding challenges related to content delivery over the internet, including technical, legal, regulatory, security, business and commercial issues.”

P2P'08: Last day, the day after the social event...

Today, I'd like to report from the last day of P2P'08 (a.k.a. the day after the social event; follow me on twitter for some details) and also give some highlights from yesterday's keynote, industry session, and so on.

Yesterday's keynote from Anja Feldmann (Deutsche Telekom Laboratories) was about the relationship between ISPs and P2P, i.e., challenges and opportunities. In particular, there needs to be information exchanged between ISPs and P2P system which is a huge issue. In fact, the ISP does not want to provide information about their topology which would help to increase the performance of the P2P system. However, she proposed a compromise in a way that the P2P system shall provide a list of destination IPs (i.e., other peers to which a peer may connect) and the ISP returns a ranked list of destination IPs based on their topology. In this way locality is provided during peer selection without exposing the ISP's infrastructure details. Some people say that this might face scalability/complexity issues as such a query might easily include thousands of IPs. A paper about this can be found here.

From the industry session I'd like to pick up two talks and briefly summarize them. The first from Martin Stiemerling (NEC), "A Peer-to-Peer SIP System based on Service-Aware Transport Overlays", with the goal to provide a generic P2P and transport overlay system that is not bound to a certain usage (e.g., files, voice, video). In that sense the developed a Service-aware Adaptive Transport Overlay (SATO) which includes dynamic adaptation to changes (network, context). However, they only focus on voice and for IPTV it is a complexity issue of transcoding thousands of streams in parallel (no SVC used). This work has been done in the context of the Ambient Networks project. The second talk I'd like to mention was from Victor Souza (Ericsson) about "P2P traffic in ISP Network" that had a strong relationship to the keynote (see above). For live streaming they propose a time shift mechanism comprising content ingestion, content migration, content lookup, and content download. They have a prototype out there that allows for both VoD and live streaming. Furthermore, they propose to place media caches near the customer. However, there were strong doubts that this will be possible. Anyway, it is interestin to see that others have similar problems as we have and how they're trying to solve them.

Traditionally, P2P is about file sharing and has been extended to games, voice, audio, video (VoD, Live Streaming, IPTV) applications' space. At P2P'08 I've discovered some further application areas:

• P2P-based file system: IGOR File System, "The IGOR File System Demonstration" (Bernhard Amann, Benedikt Elser, Yaser Houri, Thomas Fuhrmann)
• P2P-based Wiki engine: Piki - A Peer-to-Peer based Wiki Engine (Patrick Mukherjee, Christof Leng, Andy Schürr)
• P2P-based power control system: "Towards P2P Technologies for the Control of Electrical Power Systems" (Kolja Eger, Christoph Gerdes, Sebnem Öztunali)

… it seems that becoming "P2P-based" becomes the (new) trend ;-)

Next year, the conference will take place in Seattle.

P2P'08: Engineering Peer-to-Peer Systems

"Engineering = Technology + Economics" H. Schulzrinne from Columbia University said when he opened his keynote at P2P'08 in Aachen, Germany. Well, I guess most of the scientists take care only about the former and ignore the latter. In his talk he clearly outlined the increase of energy and bandwidth consumption when using P2P systems. He also quantified that but, honestly, most of the numbers were related to the US and in Europe these numbers are probably lower. However, he concluded that from an economics' point of view "P2P system shall be deployed only in Alaska and Scandinavia (and only during Winter period)" ;-)

He then defined a P2P system as follows (see also figure on the left side): (1) peers act as client and server, (2) they provide computational/storage resources to other peers, and (3) they're self-organizing and scaling. The latest property is the most important one as (1) and (2) is also true for several proxies out there...

After that he discussed another topic that sounds curious but comes back to the energy issue as mentioned above...

Finally, he also presented some technical details - I hope the slides will be provided through the Web site - and he concluded that we "need more work on diagnostics and management", little work has been done related to services such as transcoding, and standardization is ongoing within IETF.

P2P'08: Video Search & Playback in Zero-Server P2P Systems

In yesterday's tutorial at P2P'08 I learned something about zero-server P2P systems, especially related to video search and playback. Unfortunately, the BBC guy did not show up and so the audience was directly confronted with the reality: standardization in DVB ;-) However, the speaker gave a good overview how P2P systems have been used in the past and especially about the trial for the European Song Contest. Next, we got input from the industry representing the consumer electronics. They want to put a P2P engine into a settop box and raised a few very interesting issues, mainly related to CPU and memory constraints (e.g., only 4 MB of free memory for metadata or so). Also, a timely standardozation is required in order to bring interoperable products on the market. The main part of the tutorial comprises an overview about Tribler, the P2P engine also used in P2P-Next project. One interesting talk was identifying the differences between traditional file sharing, VoD, and "live" streaming. However, I always wonder how muh "live" a P2P live video stream is? Just one keyword: delay!

More to come soon ...

STreamingDay 2008, Sep. 2, Parma, Italy

At this year’s StreamingDay – kindly hosted by University of Parma – we saw a lot of interesting presentations related to video coding (AVC, reconfigurable coding, etc.), streaming, adaptation, cross-layer optimizations, peer-to-peer, robustness, and quality monitoring (see the final program for details). With this blog entry I’d like to briefly summarize the highlights of this event.

The first session was extremely hardware-oriented where they’re aiming at efficient multimedia implementations (i.e., mainly AVC encoding but also general purpose signal processing) for the ST240 VLIW processor. The overall goal is to allow for real-time encoding of HD (1080p) content which opens the door to a punch of applications.

The second session was dedicated to streaming, adaptation, and optimization issues. Michael presented our joint work (with ST) in this session (see picture). Overall, the feedback was quite good but the overhead issue is still omnipresent. In any case, the optimized reference software implementation (note: of SVC’s bitstream extractor) provides the lower boundary of what can be practically achieved.

The last session was related P2P, robustness, and quality monitoring. The P2P presentation was a good overview of P2P systems for live video streaming. In this work the authors proposed a combination of MDC and SVC for streaming of live video in P2P systems which results in higher coding efficiency and robustness but at which costs (performance and overhead)?

Conclusion: the STreamingDay is a very nice event organized by ST and related universities and you may hear a lot about the newest trends regarding streaming, coding, and applications, also from an industry perspective... hope to see you next year.