The demand for electricity related to Information and Communications Technologies is constantly growing and significantly contributes to the increase in global greenhouse gas emissions. To reduce this harmful growth, it is necessary to address this problem from different perspectives. Among these is changing the computing scale, such as migrating, if possible, algorithms and processes to the most energy efficient resources. In this context, this paper explores the possibility of running scientific and engineering programs on personal computers and compares the obtained power efficiency on these systems with that of mainframe computers and even supercomputers. Anecdotally, this paper also shows how the power efficiency obtained for the same workloads on personal computers is similar to that obtained on supercomputers included in the Green500 ranking.
Multi-population models based on evolutionary algorithms have been successfully applied to countless feature selection problems. One of the most relevant aspects is the initialization of the individuals that form the initial population, where any bias should be avoided. Usually, the initial individuals are randomly generated with the aid of the pf hyperparameter (the selection probability for any feature when generating an individual of the initial population). pf is usually set to 0.5 (the same chance of selecting or not any feature). However, different values for this hyperparameter may influence the results finally found and the execution time and energy consumed by the algorithm. This paper analyzes the behavior of a multi-objective and multi-population procedure when setting pf to different values for several datasets. The results show that there is an optimal value of pf for each dataset that speeds up the process without affecting the quality of the solutions, in addition to reducing the size of the final subset of features in some cases.