METHOD OF ORDERING OPERATIONS

An embodiment of a method of ordering operations begins with a first step of a computing entity storing a training data set, which includes previously executed operation pairs, attribute data, and relative cost data. In a second step, the computing entity forms a predictor. In a third step, the computing entity applies the predictor to an operation pair of first and second to-be-executed operations to determine a predicted relative cost between the first and second to-be-executed operations. Another embodiment of the method comprises the first step, the second step, an alternative third step, and a fourth step. In the alternative third step, the computing entity applies the predictor to to-be-executed operations in operation pairs to determine a predicted relative cost between first and second operations for each operation pair. In the fourth step, the computing entity employs the operation pairs and the predicted relative costs to sort the to-be-executed operations.