DISTRIBUTED DATA PROCESSING APPARATUS, DISTRIBUTED DATA PROCESSING METHOD, AND DISTRIBUTED DATA PROCESSING PROGRAM

摘要

PROBLEM TO BE SOLVED: To highly efficiently employ computer resources in distributed data processing for executing MapReduce.SOLUTION: A distributed data processing apparatus 10 for processing data distributed by MapReduce by a plurality of processes, calculates respective processing loads of the processes for respective data divided by prescribed sections, when the processing by Reduce function is performed by the processes. The distributed data processing apparatus 10, using the calculated processing loads of the respective data, determines allocations of the data to the respective processes such that execution times of the respective processes for executing the processing of the Reduce function are standardized.