chapter  20
30 Pages

Memetic Algorithms for Energy-Aware Computation and Communications Optimization in Computing Clusters

WithJohnatan E. Pecero, Bernabé Dorronsoro, Mateusz Guzek, Pascal Bouvry

Recently, many researchers areworkingon thedesignof new schedulers for theminimizationof the energy required to compute the assigned tasks, in addition to the minimization of the time when the latest task is finished (called themakespan time). Some examples of such techniques are the use of dynamic voltage and frequency scaling (or DVFS) to reduce the processors’ frequency for avoiding idle times [18,19,29], to maximize simultaneously the savings of different providers in the grid, instead of the total power saving [33], the use of dynamic power management (DPM) to reduce the processors’ consumption [5], or maximizing the energy saved in communications [12], between others.