Parallel Programming in Computing Undergraduate Courses: a Systematic Mapping of the Literature

Abstract

Due to the current scenario in which multi-core architectures are predominant in most personal computers and servers, the knowledge of parallel programming content becomes fundamental for computer students to develop software capable of obtaining the best performance of these architectures. Considering the importance of this context, this paper presents the results of a systematic mapping of the literature related to the teaching-learning process of parallel programming in the computing programmes in three important databases: ACM, IEEE and Science Direct. The results obtained showed that in order to solve the challenges and differences found in teaching-learning parallel programming, reorganization is necessary in the undergraduate programmes. A standard for parallel programming teaching is important. This can be established by defining where and how to insert parallelism in the courses, adopting a methodology to teach the contents of parallelism in several different courses, beginning the study in the first year. The main languages, libraries, difficulties encountered and methods of classroom and distance teaching for parallel programming are presented in this paper. Distance learning is still little explored in this area of knowledge, but it can support the teaching and study of these contents.