Applying database replication to multi-player online games

Multi-player Online Games (MOGs) have emerged as popular data intensive applications in recent years. Being used by many players simultaneously, they require a high degree of fault tolerance, scalability and performance. In this paper we analyze how database replication can be used in MOGs to achieve these goals. In data replication, clients can read data from any database replica while updates have to be executed at all available replicas. Thus, reads can be distributed among the replicas leading to reduced response time and scalability. Furthermore, the system is fault-tolerant as long as a replica is available. However, we are not aware of any previous study on the application of database replication to MOG. In this paper, we present a system, MiddleSIR, which provides database replication support. We illustrate different replication protocols implemented in the system along an example, explaining how data consistency and fault tolerance can be achieved. From there, we design a small multi-player typing game to demonstrate how to apply database replication to MOG. We will discuss how different replication protocols affect the semantics of the game. Our experiments show that database replication can provide good scalability and performance in both Local Area Networks (LAN) and Wide Area Networks (WAN).

[1]  Stuart Cheshire,et al.  Latency and the quest for interactivity , 1996 .

[2]  Jouni Smed,et al.  Realizing bullet time effect in multiplayer games with local perception filters , 2004, NetGames '04.

[3]  Jouni Smed,et al.  Realizing the bullet time effect in multiplayer games with local perception filters , 2005, Comput. Networks.

[4]  Dennis Shasha,et al.  The dangers of replication and a solution , 1996, SIGMOD '96.

[5]  André Schiper,et al.  Comparison of database replication techniques based on total order broadcast , 2005, IEEE Transactions on Knowledge and Data Engineering.

[6]  Honghui Lu,et al.  Peer-to-peer support for massively multiplayer games , 2004, IEEE INFOCOM 2004.

[7]  P.M. Sharkey,et al.  A local perception filter for distributed virtual environments , 1998, Proceedings. IEEE 1998 Virtual Reality Annual International Symposium (Cat. No.98CB36180).

[8]  Carsten Griwodz,et al.  State replication for multiplayer games , 2002, NetGames '02.

[9]  Ricardo Jiménez-Peris,et al.  Consistent Data Replication: Is It Feasible in WANs? , 2005, Euro-Par.

[10]  Gustavo Alonso,et al.  Ganymed: Scalable Replication for Transactional Web Applications , 2004, Middleware.

[11]  Ian Lintault.  A transaction execution engine architecture for multiplayer online games , 2004, NetGames '04.

[12]  Gustavo Alonso,et al.  Improving the scalability of fault-tolerant database clusters , 2002, Proceedings 22nd International Conference on Distributed Computing Systems.

[13]  Fernando Pedone,et al.  Partial replication in the Database State Machine , 2001, Proceedings IEEE International Symposium on Network Computing and Applications. NCA 2001.

[14]  I. Scott MacKenzie,et al.  Lag as a determinant of human performance in interactive systems , 1993, INTERCHI.

[15]  Jin Chen,et al.  Locality aware dynamic load management for massively multiplayer games , 2005, PPoPP.

[16]  Ricardo Jiménez-Peris,et al.  Middleware based data replication providing snapshot isolation , 2005, SIGMOD '05.

[17]  Gustavo Alonso,et al.  MIDDLE-R: Consistent database replication at the middleware level , 2005, TOCS.

[18]  Marta Patiño-Martínez.  Consistent Database Replication at the Middleware Level , 2005 .