ABSTRACT
In distributed systems, replication is used for ensuring availability and increasing performances. However, the heavy workload of distributed systems such as web2.0 applications or Global Distribution Systems, limits the benefit of replication if its degree (i.e., the number of replicas) is not controlled. Since every replica must perform all updates eventually, there is a point beyond which adding more replicas does not increase the throughput, because every replica is saturated by applying updates. Moreover, if the replication degree exceeds the optimal threshold, the useless replica would generate an overhead due to extra communication messages. In this paper, we propose a suitable replication management solution in order to reduce useless replicas. To this end, we define two mathematical models which approximate the appropriate number of replicas to achieve a given level of performance. Moreover, we demonstrate the feasibility of our replication management model through simulation. The results expose the effectiveness of our models and their accuracy.
- F. Akal, C. Türker, H. Schek, Y. Breitbart, T. Grabs, and L. Veen. Fine-Grained Replication and Scheduling with Freshness and Correctness Guarantees. In Int. Conf. on Very Large DataBase (VLDB), 2005. Google ScholarDigital Library
- G. Antoniu, J. Deverge, and S. Monnet. How to Bring Together Fault Tolerance and Data Consistency to Enable Grid Data Sharing. Concurrency and Computation: Practice and Experience, 18(13), 2006. Google ScholarDigital Library
- J. Brevik, D. Nurmi, and R. Wolski. Automatic methods for predicting machine availability in desktop grid and peer-to-peer systems. In CCGRID '04: Proceedings of the 2004 IEEE International Symposium on Cluster Computing and the Grid, pages 190--199, Washington, DC, USA, 2004. IEEE Computer Society. Google ScholarDigital Library
- E. Cecchet, G. Candea, and A. Ailamaki. Middleware-based Database Replication: The Gaps Between Theory and Practice. SIGMOD, 2008. Google ScholarDigital Library
- R. Gallersdörfer and M. Nicola. Improving Performance in Replicated Databases through Relaxed Coherency. In VLDB 1995, Proceedings of 26th International Conference on Very Large Data Bases, Zurich, Switzerland, 1995. Google ScholarDigital Library
- S. Gançarski, H. Naacke, E. Pacitti, and P. Valduriez. The Leganet System: Freshness-aware Transaction Routing in a Database Cluster. Journal of Information Systems, 32(2), 2006. Google ScholarDigital Library
- R. Guerraoui and A. Schiper. Software-Based Replication for Fault Tolerance. IEEE Computer, 30(40), 1997. Google ScholarDigital Library
- R. Koo and S. Toueg. Checkpointing and Rollback-Recovery for Distributed Systems. IEEE Transactions on Software Engineering, 13(1), 1987. Google ScholarDigital Library
- C. Le Pape, S. Gançarski, and P. Valduriez. Refresco: Improving Query Performance Through Freshness Control in a Database Cluster. In Int. Conf. On Cooperative Information Systems (CoopIS), 2004.Google ScholarCross Ref
- M. T. Özsu and Patrick Valduriez. Principles of Distributed Database Systems. Prentice Hall, 1999. Google ScholarDigital Library
- E. Pacitti, C. Coulon, Patrick Valduriez, and T. Ozsu. Preventive Replication in a Database Cluster. Distributed and Parallel Databases, 18(3), 2005. Google ScholarDigital Library
- E. Pacitti, P. Minet, and E. Simon. Fast Algorithms for Maintaining Replica Consistency in Lazy Master Replicated Databases. Int. Conf. on Very Large DataBase (VLDB), 1999. Google ScholarDigital Library
- M. Patino-Martinez, R. Jimenez-Peres, B. Kemme, and G. Alonso. MIDDLE-R, Consistent Database Replication at the Middleware Level. ACM Transactions on Computer Systems, 28(4), 2005. Google ScholarDigital Library
- PeerSim. http://peersim.sourceforge.net/.Google Scholar
- K. Ramamritham and C. Pu. A Formal Characterization of Epsilon Serializability. IEEE Transactions on Knowledge and Data Engineering, 07(6), 1995. Google ScholarDigital Library
- U. Rohm, K. Bohm, H. Sheck, and H. Schuldt. FAS - a Freshness-Sensitive Coordination Middleware for OLAP Components. Int. Conf. on Very Large DataBase (VLDB), 2002. Google ScholarDigital Library
- I. SARR, H. Naacke, and S. Gançarski. Distributed Transaction Routing with Failure Management in a Large Scale Network. In 24 Journées de Bases de Données Avancées BDA, 2008.Google Scholar
- I. SARR, H. Naacke, and S. Gançarski. DTR: Distributed Transaction Routing in a Large Scale Network. In VECPAR'08 Workshop on High-Performance Data Management in Grid Environments (selected papers), 2008. Google ScholarDigital Library
- F. B. Schneider. Implementing Fault-tolerant Services using the State Machine Approach: A tutorial. Technical report, ACM Computing surveys, 1990.Google Scholar
Index Terms
- Database replication in large scale systems: optimizing the number of replicas
Comments