Divisible Loads Scheduling in Hierarchical Memory Systems with Time and Energy Constraints

In this paper we consider scheduling distributed divisible computations in systems with hierarchical memory for energy and time performance criteria. Hierarchical memory allows to conduct computations on big data sets using out-of-core processing instead of coercing application data t into core storage. However, out-of-core computations are more costly both in time and energy. A model for scheduling divisible loads under time and energy criteria is introduced. Two types of scheduling algorithms are proposed and evaluated: a single-installment algorithm which builds optimum schedules but may use out-of-core storage, and a set of multi-installment algorithms which use limited memory but require more communications.