A method of managing computing resources and reducing execution time. The method includes: receiving, by a computing resource scheduler, a request to calculate a first value based on a first input data set; initiating, by the computing resource scheduler and on a plurality of computing resources, parallel execution of a plurality of processes to calculate the first value, where each of the plurality of processes is configured to independently calculate the first value based on the first data set; receiving, by the computing resource scheduler, the first value from a first winning process of the plurality of processes; and freeing the plurality of computing resources by aborting a losing process of the plurality of processes in response to receiving the first value from the first winning process.