Data processing apparatus, data processing method, and large-scale data processing program

摘要

PROBLEM TO BE SOLVED: To provide a data processing device capable of surely executing a query even in a restricted resource environment, in a data virtualization system.SOLUTION: A data processing device comprises: a statistical information part 20; a bucket information collection part 21; a histogram creation part 22; a query evaluation part 23; an intermediate histogram generation part 24; a use memory size calculation part 25; a division range generation part 26; and a query division part 27, in a data virtualization system. A histogram of output data for every processing of a query which is submitted by a user, is created based on a histogram for every attribute on data which is a processing object, then, a use memory size of bucket of output with respect to bucket of input is calculated, thereby a division range in which deviation of a value of output data is considered, is set, then generation of a division query is achieved with high accuracy.SELECTED DRAWING: Figure 1