Markus Esch

Epidemic Self-Synchronization in Complex Networks of Kuramoto Oscillators

In this article, we present and evaluate an epidemic scheme for the synchronization of coupled Kuramoto oscillators in communication networks. It addresses the problem of efficiently providing globally synchronous time epochs in complex, dynamic Peer-to-Peer network topologies. Rather than the usual model of continuously coupled nodes, a discretized version with sporadic message-based couplings to nearest neighbors is considered. This article empirically studies the emergence of coherent oscillator states for different network topologies, coupling functions, and sporadic coupling intensities. It further investigates the protocols minimum bandwidth requirements in small-world network topologies. Synchronization resilience under the effect of random perturbations is studied for two coupling variations. Finally, the potential utilization of the scheme for a local inference of global network topology characteristics is discussed.

Elastic Consistency in Decentralized Distributed Virtual Environments

Distributed virtual environments like online games or 3D virtual worlds have become very popular in the last few years and a more substantial surge in popularity and use is expected. Because of the huge number of users, such environments will undergo a change of infrastructure from a centralized to a decentralized one, making higher demands on consistency issues. This paper addresses consistency problems which may occur in distributed virtual environments and introduces the elastic consistency model to tackle these problems. According to elastic consistency, different consistency levels setting up a trade-off between consistency and responsiveness of the environment should be introduced in order to provide an efficient way of interaction on a global scale.

HyperVerse: simulation and testbed reconciled

When dealing with dynamic large-scale topologies such as those underlying Peer-to-Peer (P2P) Distributed Virtual Environments (DVEs), one inescapably reaches the point where either: a) simulations lack human behaviour and assessment or where b) practical experiments on a small scale do not yield significant results. The restrictions resulting from the separation of simulation and testbed environments hinder a comprehensive assessment and efficient development of adaptive algorithms and techniques for DVEs as they are investigated in our HyperVerse research project. In this paper, we present a hybrid evaluation system designed to combine the advantages of simulations and testbeds. The proposed infrastructure exhibits great flexibility particularly alluring in view of the multitude of potential research in the context of DVEs.

P2P-Based Avatar Interaction in Massive Multiuser Virtual Environments

The idea of the 3D Web as a global scale Distributed Virtual Environment (DVE) currently is very popular and a lot of research work is done in this field. In the course of the HyperVerse project we have developed a two-tier Peer-To-Peer (P2P) architecture as basic infrastructure for a federated, open and scalable 3D Web. Our approach relies on a concept that incorporates a loosely-structured P2P network overlay of user clients and a highly-structured overlay that connects a federation of reliable server machines constituting a reliable backbone service. A central problem when intending to build a massive virtual online environment is how avatar interaction as well as tracking and provision of avatar positions can be realized in a globally scalable manner. This paper presents a hybrid avatar interaction scheme developed for HyperVerse that incorporates the user clients and the backbone service for avatar position tracking. The backbone is used as reliable fall-back service, while the avatar tracking is handled in a pure P2P fashion whenever possible. This way backbone overload is prevented in a self-organizing manner whenever client density tends to overburden the backbone infrastructure.

The SEMPA prototype - using XAML and web services for rich interactive Peer-to-Peer applications

Currently one can see a surge of evolving technologies supporting the creation of Rich Internet Applications. All of these approaches however traditionally address the classic Client/Server scenario in which a dedicated Web Server acts as application provider for thin clients. This paper argues, that there are many scenarios in which a Peer-to-Peer approach enabling every application/process to export parts of its graphical user interface in a lightweight and interoperable way would be more desirable. This work-in-progress paper presents the lightweight middleware prototype SEMPA, proving that these scenarios can be supported by combining the readily available technologies eXtensible Application Markup Language (XAML) and Web Services.

Evaluation of the HyperVerse Avatar Management Scheme Based on the Analysis of Second Life Traces

Massive Multiuser Virtual Environments (MMVEs) and the idea of a global scale 3D Web have grown popular in recent years. While commercial precursors of such environments for the most part rely on centralized client/server architectures, it is commonly accepted that a global scale virtual online world can only be realized in a distributed fashion. Within the HyperVerse project, we have developed and recently presented a two-tier Peer-to-Peer (P2P) architecture that incorporates a loosely structured P2P overlay of user peers and a highly structured overlay of server machines constituting a reliable backbone service. In such a distributed environment, an essential question is how avatars are tracked and interconnected in order to allow mutual rendering and interaction. We have previously proposed a hybrid avatar management scheme that utilizes the backbone service for avatar tracking if necessary, but handles tracking in a P2P fashion when peers can track each other to reduce the backbone load. This paper presents a detailed performance analysis of this algorithm under a realistic scenario, using traces from a large scale MMVE called Second Life. Moreover this paper presents and evaluates an optimization for the hybrid avatar tracking scheme that can be utilized under a weaker condition.

A scale-free and self-organized P2P overlay for massive multiuser virtual environments

Massive Multiuser Virtual Environments have recently grown popular, and commercial virtual online worlds like Second Life or World of Warcraft attract a lot of attention. In this context, for the research on distributed systems, especially the idea of a 3DWeb as a global scale virtual environment is very interesting, since it poses severe technical challenges to the underlying infrastructure. It is generally accepted, that the realization of such a global scale scenario can not be realized in a traditional centralized fashion. For this reason in the course of the HyperVerse project we have developed a two-tier Peer-to-Peer (P2P) architecture as basic infrastructure for a federated and scalable 3D Web. Our approach relies on a concept, that incorporates a loosely-structured P2P overlay of user clients and an overlay that connects a federation of reliable server machines constituting a reliable backbone service. This paper proposes a self-organized and scale-free network overlay for the reliable backbone in the HyperVerse architecture. The overlay incorporates advantages of scale-free networks, self-organization and epidemic aggregation in order to tackle the severe challenges of the scenario in a fully distributed fashion without any central control.

SimCon - a simulation and visualization environment for overlay networks and large-scale applications

A substantial number of massive large-scale applications require scalable underlying network topologies. Nowadays structured Peer-to-Peer overlay networks meet these requirements very well. But there is still a need to decide which of these overlay networks is most suitable for providing the best possible performance for a certain application. This paper describes SimCon - a simulation environment for overlay networks and large-scale applications. SimCon allows the comparison of different overlay networks with respect to predefined metrics derived from requirements of the considered application. This approach allows determining which overlay network meets the needs of a given application best, which in turn is a great support for developers of large-scale applications.

Using Epidemic Hoarding to Minimize Load Delays in P2P Distributed Virtual Environments

Distributed Virtual Environments (DVEs) have grown popular in various fields of application. Apart from providing great collaborational opportunities in an immersive setting, large-scale DVEs pose severe scalability challenges. Although P2P approaches have proven to be effective for tackling many of these issues, still load delay problems remain in regions with high object or avatar density. In this article we present and evaluate a hoarding approach that is suitable to minimize such delays in P2P-based DVEs with a real-time distribution of dynamic data. The prediction of what data shall be hoarded is based on an epidemic aggregation algorithm working solely with local knowledge. Evaluation results that have been obtained using a DVE simulation environment will be presented.

Complex Structures and Collective Dynamics in Networked Systems: A Tutorial

The study of complex networks and collective dynamics occurring in biological, social and technical systems has experienced a massive surge of interest both from academia and industry. Many of the results on the mechanisms underlying the self-organized formation of complex dynamic networks in natural and man-made systems have been derived based on a statistical physics perspective. In this tutorial, we provide a basic introduction to this perspective which will help attendees to benefit from the vast literature on self-organization and self-adaptation phenomena available in the fields of network science and complex systems. We cover basic models and abstractions for the study of static complex networks as well as dynamical processes like, e.g., information diffusion, random walks, synchronization or the propagation of cascading failures. We further introduce recent advances in the study of dynamic (social) networks and demonstrate how the resulting methods can be practically applied in the engineering of self-organizing and self-adaptive distributed systems and protocols.