On correctness of scalable multi-server state replication in online games

Müller, Jens; Gößling, Andreas; Gorlatch, Sergei

doi:10.1145/1230040.1230088

Cited by 8 publications

(5 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While some authors consider the consistency of clientside part of single-server online games [7], we rather focus on the more challenging problem of concurrent update operations on global objects in multi-server environments. Previous work [5,6] to improved the correctness of the state-based approach in multi-server online games while preserving correctness. The time warp approach of [5,6] delays the delivery of state updates to ensure correctness, whereas [4] also considers the correlation of updates operations and proposes a rollback-based algorithm which improves the performance for unrelated updates.…”

Section: Conclusion and Related Workmentioning

confidence: 99%

“…Previous work [5,6] to improved the correctness of the state-based approach in multi-server online games while preserving correctness. The time warp approach of [5,6] delays the delivery of state updates to ensure correctness, whereas [4] also considers the correlation of updates operations and proposes a rollback-based algorithm which improves the performance for unrelated updates. The latter approach is similar to the update ordering approach [1] which we adopt in our paper.…”

Section: Conclusion and Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Improving the Concurrent Updates of Replicated Global Objects in Multi-Server Virtual Environments

Bartholomäus

Ploß

Gorlatch

2011

Proceedings of the 4th International ICST Conference on Simulation Tools and Techniques

Self Cite

View full text Add to dashboard Cite

We study an emerging class of high-performance virtual environments, called Real-Time Online Interactive Applications (ROIA), with such popular examples as Massively Multi-player Online Games (MMOG), interactive simulations, virtual communities, etc. ROIA must handle an enormous number of actions from geographically distributed user processes and present to each user a consistent view of the application state. We address a challenging aspect which influences the application scalability-the simultaneous access of several clients and servers to globally replicated objects. The traditional primary-copy-based state replication mechanism ensures fast read access but shows drawbacks for global objects that are concurrently updated by multiple processes. Our main contribution is to combine the traditional statebased approach with a novel operation-based update ordering approach that considers the operation's semantics and allows to use weaker consistency models than the sequential consistency enforced by the primary copy protocol. We implement our improvement of the update concurrency as the Global Object Management System (GOMS) and integrate it into the Real-Time Framework (RTF)-our platform for the high-level development and efficient execution of ROIA which has been used for implementing numerous applications like real-time online MMOG, crowd simulation, etc. Our experiments with GOMS show significant improvements regarding both the usability and the performance of the concurrent updates to replicated global objects.

show abstract

Section: Conclusion and Related Workmentioning

confidence: 99%

Section: Conclusion and Related Workmentioning

confidence: 99%

Improving the Concurrent Updates of Replicated Global Objects in Multi-Server Virtual Environments

Bartholomäus

Ploß

Gorlatch

2011

Proceedings of the 4th International ICST Conference on Simulation Tools and Techniques

Self Cite

View full text Add to dashboard Cite

show abstract

“…Several designs have been proposed that improve state consistency. Optimistic approaches, such as TimeWarp [2], rewind and recalculate state when new information is discovered; this helps correctness [3], although it is not always suitable for real-time games. A more practical technique is Local lag, which introduces a delay in local event processing, providing time for the corresponding network messages to reach their destinations [2].…”

Section: Background and Related Workmentioning

confidence: 99%

A protocol for distributed collision detection

Chen¹,

Verbrugge²

2010

2010 9th Annual Workshop on Network and Systems Support for Games

View full text Add to dashboard Cite

Abstract-Scalability of multiplayer games can be improved by client-side processing of game actions. Consistency becomes a concern, however, in the case of unpredictable but important events such as object interactions. We propose here a new motion-lock protocol for distributed collision detection and resolution. The motion-lock protocol improves performance of motion prediction by giving stations time to communicate and agree on the detected collisions. This reduces the divergence of distributed object states and post-collision trajectories. Offline and online simulation results show the motion-lock protocol results in qualitative and quantitative improvements to consistency, with negligible network impact and a minimal sacrifice in the responsiveness of player controls. Our design can be used to hide latency and reduce server load in current multiplayer online games, improving scalability and furthering fully distributed designs.

show abstract

“…The series has pursued a far more open model of development than many, with both a very open architecture allowing for extensive user-created content and the release of source code under a GPL license [15]. The combination of Quake's ubiquity as a reference model and its openness as an architecture have lead to its widespread adoption as a research testbed for traffic analyses [2,19] and distributed architectures [6,27] of multiplayer games.…”

Section: Figure 4 a Depiction Of The Simulation Setup Client C 1 Ismentioning

confidence: 99%

An Evaluation of Push-Pull Algorithms in Support of Cell-Based Interest Management

Ferscha

2007

11th IEEE International Symposium on Distributed Simulation and Real-Time Applications (DS-RT'07)

View full text Add to dashboard Cite

Several approaches for scalable interest management (IM) within real-time distributed virtual environments (DVEs) have been proposed based upon some division of the data-space in to disjoint volumes or cells. Responsibility for the entire space can then be distributed. Any such approach, however, must implement some mechanism for propagating the query and update messages around the distributed system. The efficiency of this process can greatly effect the scalability of such systems. In this paper we evaluate an adaptive approach to this problem, designed for use in cell-based systems in general.

show abstract

On correctness of scalable multi-server state replication in online games

Cited by 8 publications

References 7 publications

Improving the Concurrent Updates of Replicated Global Objects in Multi-Server Virtual Environments

Improving the Concurrent Updates of Replicated Global Objects in Multi-Server Virtual Environments

A protocol for distributed collision detection

An Evaluation of Push-Pull Algorithms in Support of Cell-Based Interest Management

Contact Info

Product

Resources

About