| 摘要 |
PROBLEM TO BE SOLVED: To reduce a communication amount in the case of successively processing a query for performing distributed processing of discontinuous data and a query for performing distributed processing of continuous data in a predetermined block, collected for each attribute of a category, so as to improve a throughput.SOLUTION: The data processing method is provided to allow a computer to execute a first query for performing distributed processing of a plurality of successively input data on the basis of a first distribution key and a second query for performing distributed processing of continuous data in a predetermined block among the plurality of data on the basis of a second distribution key. The data processing method includes procedures of: calculating a value indicating a relationship between the first distribution key for specifying the attribute of a category extracted from the syntax information of the first query and the second distribution key for specifying the attribute of an area divided in accordance with the distribution of the coordinates of the continuous data in the predetermined block; generating a query group by grouping the first query and the second query on the basis of the calculated value indicating the relationship; and performing distributed processing of the first query and the second query for each query group.SELECTED DRAWING: Figure 31 |
