The usage of Single-Board Computers (SBCs) by scientists, students, and hobbyists is soaring to new heights due to their affordability, versatility, low energy consumption, and increasing community support. On the other hand, the benefits of containerization have fostered its acceptance and deployment. Due to the growing interest in this latter technology, the community has developed several container engines. In this paper, three Raspberry Pi models were selected and empirically assessed when running containers using leading container engines (Docker, Podman, and LXC/LXD). The paper starts by benchmarking the processing power observed at the level of the three container technologies for the computation of π, when varying the SBC model and the number of concurrent containers. Then, the initial life of containers is assessed by obtaining their creation, start, and run times. Finally, the network performance between containers is evaluated when using WiFi and Ethernet. In these experiments, Docker and Podman exhibited similar performances (they are focused on "application containers"). LXC/LXD was generally slower since it is aimed at "system containers". The paper also mentions other important factors intended to further guide users in selecting SBCs or container engines for their use cases.
No takes yet. Share an insight, caveat, or question.
Gamess et al. (2024) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: