BLOCK ACCESS ESTIMATION FOR CLUSTERED DATA USING A FINITE LRU BUFFER

Data access cost evaluation is fundamental in the design and managemen t of database systems. When some data items have duplicates, a cluster ing effect which can heavily influence access costs is observed. The a vailability of a finite amount of buffer memory in real systems has an even more dramatic impact. In this paper a comprehensive cost model f or clustered data retrieval by an index using a finite buffer is prese nted. Our approach combines and extends previous models based either o n finite buffer or on uniform data clustering assumptions. The computa tional cost of the formulas we propose in this work is independent of the data size or of the query cardinality and need only a single stati stics per search key, the clustering factor, to be maintained by the s ystem. The predictive power and the accuracy of the model are shown in comparison with actual costs resulting from simulations.