Modeling and analyzing the energy consumption of fork-join-based task parallel programs

摘要

Because of environmental and monetary concerns, it is increasingly important to reduce the energyrnconsumption in all areas, including parallel and high performance computing. In this article, we propose anrnapproach to reduce the energy consumption needed for the execution of a set of tasks computed in parallelrnin a fork-join fashion. The approach consists of an analytical model for the energy consumption of a parallelrncomputation in fork-join form on dynamic voltage frequency scaling processors, a theoretical specificationrnof an energy-optimal frequency-scaled state, and the energy minimization by computing optimal scalingrnfactors. For larger numbers of tasks, the approach is extended by scheduling algorithms, which exploit thernanalytical result and aim at a reduction of the energy. Energy measurements of a complex numerical methodrnand the SPEC CPU2006 benchmarks as well as simulations for a large number of randomly generated tasksrnillustrate and validate the energy modeling, the minimization, and the scheduling results.