A computer-implemented method is disclosed for efficiently processing a large number of records. In the method, a computer system may obtain a plurality of records and count the number of records thereof corresponding to each block of a plurality of blocks. The computer system may also identify a plurality of partitions corresponding to selected blocks of the plurality of blocks. Each partition of the plurality of partitions may be substantially uniform in processing time. The computer system may then distribute a workload associated with a block or partition to each node of a plurality of nodes contained within the computer system. Each node may then process the block or partition in parallel such that each node completes the processing within a selected period of time.