Abstract: In a distributed computer system, a data set having a plurality of data rows is extended by adding an additional column to the data set and assigning a persistent partition ID to each of the plurality of data rows in the additional column. The extended data set can then be queried by a plurality of queries, each of which has conditions that select the same rows from the data set and each of which has different partition ID information. The partition ID information forms a different result set from the selected rows so that the result set of each query forms a non-overlapping, transferable result set partition. Together, the result set partitions form the complete result set. The result set partitions can be transferred in parallel through the computer system to increase the system throughput.