Competitive Analysis of Caching in Distributed Databases

This paper makes two contributions. First, we introduce a model for evaluating the performanceof data allocation and replication algorithms in distributed databases. The model is comprehen-sive in the sense that it accounts for I/O cost, for communication cost, and, because of reliabilityconsiderations, for limits on the minimum number of copies of the object. The model captures ex-isting replica-management algorithms, such as read-one-write-all, quorum-consensus, etc. Thesealgorithms are static in the sense that, in the absence of failures, the copies of each object areallocated to a xed set of processors.In modern distributed databases, particularly in mobile computing environments, processors willdynamically store objects in their local database and will relinquish them. Therefore, as a secondcontribution of this paper, we introduce an algorithm for automatic dynamic allocation of repli-cas to processors. Then, using the new model, we compare the performance of the traditionalread-one-write-all static allocation algorithm, to the performance of the dynamic allocation algo-rithm. As a result, we obtain the relationship between the communication cost and I/O cost forwhich static allocation is superior to dynamic allocation, and the relationships for which dynamicallocation is superior.