Analysis of Replication in Distributed Database

Geographically distributed database systems have re- ceived growing interest in recent years. In this paper, we develop an approximate analytical model to study the tradeoffs of replicating data in a distributed database environment. Several concurrency control protocols are considered including pessimistic, optimistic, and semi- optimistic protocols. The approximate analysis captures the effect of the protocol on hardware resource contention and data contention. The accuracy of the approximation is validated through detailed simula- tions. We find that the benefit of replicating data and the optimal num- ber cf replicates are sensitive to the concurrency control protocol. Un- der the optimistic and semi-optimistic protocols, replications can significantly improve response time with an additional MIPS require- ment to maintain consistency among the replicates. The optimal degree of replication is further affected by the transaction mix (e.g., the frac- tion of read-only transactions), the communications delay and over- head, the number of distributed sites, and the available MIPS. Sensi- tivity analyses have been carried out to examine how the optimal degree of replication changes with respect to these factors.