A Preliminary Study of the Energy Impact of Software in Raspberry Pi devices

Kamar Kesrouani (1) , Houssam Kanso (1) , Adel Noureddine (1)

1 LIUPPA - Laboratoire Informatique de l'Université de Pau et des Pays de l'Adour

Nowadays, the use of IoT devices is essential in most sectors along with a rising concern regarding their energy consumption. Therefore, an accurate estimation of the energy consumption of such devices is required for energy-efficient improvements. This paper presents Energy Measurement Model (EMM), an energy estimation tool for Raspberry Pi 3 Model B+. It is a software-based model used to estimate the local energy consumption of a device by taking into consideration CPU utilization. The error rate of our model averages at 1.25%. Using this model, we study the energy impact of a set of algorithms, programming languages and compilers.

Mots clés: Index Terms-Connected Objects; Raspberry Pi; IoT; Energy Consumption; Energy Efficiency; Software Engineering

Domaines: Génie logiciel [cs.SE]; Architectures Matérielles [cs.AR]; Systèmes embarqués

Kamar Kesrouani, Houssam Kanso, Adel Noureddine. A Preliminary Study of the Energy Impact of Software in Raspberry Pi devices. 29th IEEE International Conference on Enabling Technologies: Infrastructure for Collaborative Enterprises, 2020, Bayonne, France. ⟨hal-02936861⟩