Georgio Lucarelli - Energy-aware scheduling algorithms

Due to the increasing need for dealing with the energy consumption in computer systems, new low energy architectures have been developed that support several low power features such as multiple power states of memory banks, bank/row specific activation, and the dynamic voltage scaling of processors. These features force to take appropriate decisions at higher design levels. Thus, new technologies generate new algorithmic questions, as by answering them is likely to have a significant impact in reducing the energy consumption in computer systems. In this context it appeals very interesting to address these problems as scheduling problems, with the objective of minimizing the energy consumption. In this talk, we consider the energy minimization problem of scheduling a set of n jobs on m parallel speed-scalable processors. This model captures the intuitive idea that the faster a processor works the more energy it consumes. We present different algorithmic approaches that lead to optimal or approximate solutions for several machine environments.