Richard Sueselbeck

Bio: Richard Sueselbeck is an academic researcher from University of Mannheim. The author has contributed to research in topics: Virtual machine & Computer cluster. The author has an hindex of 1, co-authored 1 publications receiving 16 citations.

Adaptive Update Propagation for Low-Latency Massively Multi-User Virtual Environments

Abstract: Massively Multi-User Virtual Environments (MMVEs) are highly interactive systems. They require the propagation of state updates to users with little delay. In this paper we propose a novel update propagation approach for MMVEs that enables such low-latency propagation while offering the scalability needed to support MMVEs with massive user numbers. Our approach combines peer-based and server- based update propagation into a hybrid system. It adapts itself dynamically to the available system resources and the current situation in the virtual world. We describe our approach in detail, evaluate it and discuss further steps towards a low-latency update propagation system. I. INTRODUCTION Massively Multi-User Virtual Environments (MMVEs) al- low thousands of users worldwide to interact with each other in a common virtual environment in real-time. Such systems are highly interactive. Users expect the virtual environment to react to their actions immediately. Thus, updates of the state of an MMVE must be propagated with little delay. To do so, a suitable update propagation subsystem is needed, which ensures that all users receive all updates they require in time. Current MMVEs are based on client/server-architectures. A central server cluster collects all user actions and sends update messages containing an updated state of the virtual world to all clients. However, this approach introduces an additional delay for update propagation since updates are first sent to the server and then back to the clients. Lower delays are possible by using directly connected peer computers. Peers send updates directly to each other. Unfortunately, due to the limited upload bandwidth of the peers, this approach is not scalable enough to be usable for MMVEs with massive user numbers. In this paper we propose an update propagation subsystem for MMVEs that combines both of these approaches into a hybrid system. Based on a central server we dynamically shift the responsibility for update propagation between the server and individual peers such that the resulting propagation delay is minimized. At the same time we maintain the scalability of the server-based architecture. Our approach is based on the notion of so-called Areas of Propagation (AoP). An AoP determines the area in the virtual world for which a peer can distribute updates directly. That is, if an update occurs on a peer, the peer first checks if the update will influence only peers that are in its own AoP. If so, it delivers the update to them directly. Otherwise, it notifies the server. The size of a peer's AoP is selected dynamically depending on the peer's available bandwidth and the density of peers in the virtual world. This allows to communicating directly with the maximum possible number of peers in the virtual environment, while maintaining the scalability of previous approaches. Our contributions in this paper are as follows: first, we introduce AoPs, our concept to integrate peer-to-peer-based and server-based update propagation. Second, we present an approach to create and maintain AoPs at runtime. Third, we show how to use AoPs to create a scalable and low delay update propagation system. Finally, we give a short evaluation of our approach and compare it to pure P2P-based and pure server-based update propagation. Our work is part of the peers@play project (3), a cooperative project of the Universities of Mannheim, Duisburg-Essen and Hannover to develop protocols and algorithms for highly scalable and interactive MMVEs. We build upon existing functionality from this project to realize our approach. Where necessary, we describe this functionality briefly.

Peer-to-peer support for low-latency Massively Multiplayer Online Games in the cloud

Abstract: Cloud gaming has recently been proposed as an alternative to traditional video game distribution. With this approach, the entire game is stored, run and rendered on a remote server. Player input is forwarded to the server via the Internet and the game's output is returned as a video stream. This adds network delay, which can negatively impact the gameplay. The delay is acceptable as long as the user is located geographically close to the cloud servers. However, for Massively Multiplayer Online Games (MMOGs), this delay is added on top of the existing delay between MMOG client and server. As MMOGs are highly delay-sensitive, this can significantly degrade their playability. To deal with this issue, we propose to use peer-to-peer techniques to distribute the MMOG server functionality and place it at the cloud server centers. This allows us to reduce the additional delay introduced by running the MMOG clients in the cloud.

Efficient bandwidth estimation for peer-to-peer systems

Abstract: Many peer-to-peer (P2P) systems require accurate information about their peer's available bandwidth, e.g., for load balancing. Determining this information is difficult, as a suitable approach must address two challenges. First, it must be able to deal with fluctuating bandwidth. Second, it must incur low overhead to prevent interference with the operation of the P2P system. In this paper we present an approach to estimate the available bandwidth of peers in a P2P system, based on a combination of traffic observation and the strategic injection of traffic into the system. We evaluate our approach and show that it is accurate and responsive in settings with variable bandwidth while resulting in limited interference with the system.

SPEX: scalable spatial publish/subscribe for distributed virtual worlds without borders

Abstract: Existing architectures designed to host large-scale virtual environments (VEs) use a variety of approaches, but they often limit the interaction range with other users or with the VE. How densely users can populate a given region is also limited by the hosting machine's CPU or bandwidth resources. We are motivated to remove such restrictions and present SPEX, an infrastructure that supports scalable spatial publish/subscribe for VE applications. SPEX is scalable and fault-tolerant, with adaptive load balancing and low latency as its key features. It is designed for the state and overlay management in VEs with many concurrent users. We evaluate a practical SPEX implementation within Amazon's EC2 Cloud and present a feasible approach to supporting 750 users across a continent with low latency, opening the possibility for hosting fast-paced games (e.g., first-person shooters) or applications on a large-scale.

Consistency management for interactive peer-to-peer-based systems

Abstract: Consistency is a crucial requirement for Massively Multiuser Virtual Environments (MMVEs). Such systems provide virtual worlds in which thousands of users can interact in realtime. To realize a consistent world view for all users, a flexible consistency management is required which balances the responsiveness and the level of consistency in the system. In this paper we present an approach for a consistency management for peer-to-peer-based MMVEs. The approach identifies users which actually interact with each other in the virtual world, groups them in consistency sessions and synchronizes all users in the session according to a synchronization protocol which is determined at runtime. We discuss the concept of consistency sessions, present algorithms to create and maintain sessions, and analyze what features are still missing and must be added in the future.