Sustainable development in the area of Information Technology (IT) has become a reality and a necessity not only in IT centers, with the reuse and recycling of equipment, but also in large industries with the use of good sustainable production practices and reverse logistics. The development of computer clusters became a sustainable option in the reuse of computers that would be discarded and an incentive in the teaching and learning process of Parallel Programming. This work presents an analysis carried out among the teachers and students of the Federal Institute of Pará -Campus Abaetetuba, in the Amazon region, on the need to insert contents related to the area of High-Performance Computing in the curricular matrices of the technical courses of Informatics of the Institution. The survey showed that 71% of the participating teachers would like to have a cluster in the Institution, while 56% of the students would like to have content of this nature in the curricular matrices of their courses.

The use of heterogeneous computing (CPU and GPU) in general purpose application processing has evolved exponentially in recent years. This type of architecture aims to accelerate the achievement of results, which contributes to the reduction in processing time. For this reason, these systems have also been used in applications that need to recognize some type of pattern, such as facial recognition. Facial recognition systems have gained notoriety in recent years because they are more accurate and non-invasive in the process of identifying people previously registered in a database. With this objective, this work uses two parallel libraries, TensorFlow and PyTorch, in order to evaluate the reduction of processing time in a facial recognition application that uses an Artificial Neural Network. The results were promising with the reduction of processing time up to three times when compared to sequential processing time.