Charilaos Stais

A BitTorrent module for the OMNeT++ simulator

In the past few years numerous P2P file-sharing and content distribution systems have been designed, implemented, and evaluated via simulations, real world measurements, and mathematical analysis. Yet, only few of them have stood the test of time and gained wide user acceptance. BitTorrent is the one that holds the lions share among them and the reasons behind its success have been studied to a great extent with interesting results. Nevertheless, even though P2P content distribution remains one of the most active research areas, little progress has been made towards the study of the BitTorrent protocol (and its variations), in a fully controllable and realistic simulation environment. In this paper we describe and analyze a full-featured and extensible implementation of BitTorrent for the OMNeT++ simulation platform. Moreover, since we aim at realistic simulations, we present our enhancements on a popular conversion tool for practical Internet topologies, as well as our churn generator that is based on the analysis of real BitTorrent traces. Finally, we set forth the results from the evaluation of our prototype implementation regarding resource demands under different simulation scenarios.

Networked music performance over information-centric networks

Information-centric networking (ICN) constitutes an alternative to the conventional, IP-based, internetworking, with information itself being identified rather than the host where it resides. This approach introduces powerful tools and operations for content delivery, such as native support for multicast. Exploiting this native multicast capability is a very promising approach for multimedia applications such as Networked Music Performance (NMP), where a set of musicians located in different places wish to perform together in real time. While conferencing applications traditionally rely on a Multipoint Conferencing Unit (MCU) that receives media streams from each participant and then retransmits a mixed stream to each one, in NMP we would prefer direct communication between the participants, so as to reduce transmission delays and allow each participant to mix the incoming media streams in the desired manner. In this paper we introduce an ICN-based NMP application exploiting native multicast, and compare its performance with both MCU and non-MCU NMP variants, using both unicast and multicast.

Towards an error control scheme for a publish/subscribe network

Many proposals for the next generation of the Internet suggest moving from an end-point oriented to an information-centric oriented architecture. Many of these proposals are based on the publish/subscribe paradigm, which lends itself naturally to native multicast support, a key factor for efficient content distribution. However, the design of efficient reliable transport protocols for multicast is a largely open problem, due to the problem of feedback implosion towards the sender as group size grows. In this paper we propose a hierarchical retransmission-based error control scheme for a native publish/subscribe internetwork. We compare our protocol with similar approaches proposed for IP multicast and evaluate its performance against IP multicast with unicast-based error control.

Supporting mobility in a publish subscribe internetwork architecture

Information-Centric Networking (ICN) is constantly gaining momentum within the Future Internet research community. In the PURSUIT research project we are developing a clean-slate Pub/Sub Internetworking (PSI or Ψ) approach with integrated seamless mobility support. The novel ICN mechanisms supported in Ψ, along with smartly placed in-network caches, enable the architecture to handle both mobile and fixed devices in a uniform way. This paper presents a blueprint for optimizing mobility support in Ψ without modifications to the architecture or add-on solutions. We demonstrate a micro-mobility scenario that describes the functionality of Ψs core components in supporting mobility and then sketch our plans for future work and a proper assessment of these designs.

A reliable multicast transport protocol for information-centric networks

In the past few years, many researchers have argued that the Internet should transit from its traditional endpoint-centric architecture to an information-centric paradigm. One of the advantages of the information-centric model is that the network can easily aggregate requests for the same content and serve them via multicast. Indeed, most information-centric architectures proposed to date offer native support for multicast, promising a vast improvement in the efficiency of content distribution. However, designing efficient reliable transport protocols for multicast is a largely open issue, due to the problem of feedback implosion towards the sender as group size grows. In this paper we propose RMTPSI, a retransmission-based reliable error control protocol for multicast communication designed specifically for information-centric networks. We compare RMTPSI with existing approaches proposed for IP multicast and evaluate its performance via simulation, showing that our approach leads to more efficient content distribution and error recovery than previous solutions.

Sink controlled reliable transport for disaster recovery

We present a reliable transport layer protocol for sensor networks, targeting disaster recovery applications where human or robotic rescuers try to gather information from a possibly fragmented sensor network by moving through the disaster area. The mobility of the information sink means that the protocol must quickly adapt to a constantly changing view of the network, where connections and disconnections are the norm. Our protocol is purely sink driven, that is, the sink controls congestion by rate limiting the sensors, choosing how to assign the available bandwidth to different sensor types and deciding on the level of reliability to be achieved. In addition, our protocol operates at the application layer with minimal requirements from lower layers, allowing its integration with a disaster recovery application that will set its parameters depending on the disaster scenario. As a result, our protocol allows simple and inexpensive fixed sensors to be combined with expensive but reusable mobile equipment for disaster recovery purposes.

A comparison of streaming extensions to BitTorrent

Peer-to-peer (P2P) protocols have been widely accepted by users and operators alike as efficient mechanisms for non real-time content distribution. It is therefore reasonable to extend these protocols to also handle more demanding applications, such as multimedia streaming. Many researchers have proposed modifications to the well known BitTorrent protocol in order to adapt it to the needs of such applications. In this paper we present findings from our experiments with three proposals for the integration of multimedia streaming into BitTorrent: the fixed-size window approach, the high-priority set approach and the stretching window approach. We evaluate these proposals under identical circumstances using our detailed packet-level BitTorrent simulator, showing that while all approaches are reasonably capable of supporting multimedia streaming, their different design choices have a pronounced effect on their streaming performance.

Demonstrating an Information-Centric Network in an International Testbed

Information-Centric Networking (ICN) has increasingly been attracting attention by the research community. In ICN the center of attention becomes the information itself and not the endpoints as in today’s IP networks. In this demonstration we present applications that we developed as proof of concepts for our ICN approach. A video streaming as well as a voice and a HTTP over publish/subscribe application that run on top of our ICN prototype will be demonstrated running in an international testbed.

RT-SENMOS: Sink-Driven Congestion and Error Control for Sensor Networks

A very common problem in the sensor networks area, where large numbers of sensors transmit data to a single sink node, is reliable transport. As sensors become cheaper and more powerful, sensor networks face the challenge of controlling all these devices in a efficient manner, which naturally leads to a sink-driven transport scheme. In this paper we propose RT-SENMOS, a sink-driven reliable transport protocol for sensor networks which places the responsibility for transmission rate allocation, congestion avoidance and error control to the sink. RT-SENMOS is intended to be integrated within a specific application which will set its operating parameters depending on its needs, operating on top of UDP/IP to avoid the need for kernel/superuser access. As RT-SENMOS is fully sink-controlled, it enables the use of simple and inexpensive fixed sensors, which offload all protocol intelligence to a more expensive but reusable sink. We present the design of the protocol, comparing it with a similar approach, called RCRT, and evaluate its performance using a real implementation.

Demonstrating usage diversity over an information-centric network

Information-Centric Networking (ICN) is becoming increasingly popular: various architectures have been proposed to transform todays host-centric networks to information-centric ones. An important question that begs to be answered is whether the ICN paradigm can adequately support various types of existing network usage, as well as set the foundations for new application types. In this demonstration, we exhibit four different communication models via corresponding applications supported by our architecture. All applications are realized through the publish/subscribe service model exported by our ICN nodes.