Analyzing programming languages' energy consumption

Abstract

Motivation: The energy efficiency of it-related products, from the software perspective, has gained vast popularity the recent years and paved a new emerging research field. However, there is limited number of research works regarding the energy consumption of relatively small programming tasks. This knowledge is critical to be known especially in cases where millions of small tasks are running in parallel on multiple devices all around the globe. Goal: In this preliminary study, we aim to identify energy implications of small, independent tasks developed in different programming languages; compiled, semi-compiled, and interpreted ones. Method: To achieve our purpose, we collected, refined, compared, and analyzed a number of implemented tasks from Rosetta Code, that is a publicly available Repository for programming chrestomathy. Results: Our analysis shows that among compiled programming languages such as C, C++, Java, and Go offer the highest energy efficiency for all of our tested tasks compared to C#, vb.net, and Rust. Regarding interpreted programming languages php, Ruby, and JavaScript exhibit the most energy savings compared to Swift, R, Perl, and Python.