"Exploring the Virtualization Potential of Raspberry Pi Devices with Pr" by Deuntae Winston
 
Exploring the Virtualization Potential of Raspberry Pi Devices with Proxmox VE

Exploring the Virtualization Potential of Raspberry Pi Devices with Proxmox VE

Date

2-19-2025

Faculty Mentor

Eric Gamess, Mathematics, Computing & Information Science; Sabin Banjara, Mathematics, Computing & Information Science

Files

Submission Type

Conference Proceeding

Location

8:45-8:55 am | Houston Cole Library, 11th Floor

Description

Please note: no video is available for this presentation.

Single-board computers (SBCs) are becoming increasingly popular for all types of applications due to their versatility, affordability, low power consumption, growing computing capabilities, and active community support. They were originally marketed for Internet of Things (IoT) applications, DIY projects, and educational purposes. However, over the years, the development of more powerful models has opened the way for resource-intensive applications such as virtualization. This work explores the feasibility and limitations of using the Raspberry Pi 4 Model B (RPi 4B) and the Raspberry Pi 5 (RPi 5) to create affordable virtual environments with Proxmox VE. Proxmox VE is an open-source virtualization platform for the management of virtual machines (VMs), containers, and storage resources. The study considers various memory configurations and storage mediums for the evaluated SBCs, including microSD cards, SATA SSDs connected through USB 3.0, and NVMe SSDs connected through PCIe x1. To assess and analyze the potential of the proposed virtual environment, exhaustive experiments were conducted, focusing on VM cloning time, startup efficiency, scalability, live migration, and network performance.

Our experiments showed that both Raspberry Pis (RPi 4B and RPi 5) can be utilized in virtual environments, albeit with a limited number of VMs, primarily due to memory constraints. The SBCs with 4 GB of RAM could run up to six basic Debian VMs, while the 8 GB versions supported a maximum of fifteen VMs. The RPi 5, with its upgraded hardware, significantly outperformed the RPi 4B, making it a better choice for projects requiring higher processing power, if the slight increase in cost fits within the budget. Moreover, the PCIe x1 interface available in the RPi 5 (but not in the RPi 4B) proved to be a key advantage in all the experiments performed in this work, offering a much higher data transfer rate. For the RPi 4B, storage access emerged as the primary bottleneck. In contrast, with the RPi 5 equipped with an NVMe SSD connected through PCIe x1, the network became the most common limitation, as storage access speeds far exceeded network transfer rates.

Keywords

student research, computing

Rights

This content is the property of Jacksonville State University and is intended for non-commercial use. Video and images may be copied for personal use, research, teaching or any "fair use" as defined by copyright law. Users are asked to acknowledge Jacksonville State University. For more information, please contact digitalcommons@jsu.edu.

Disciplines

Computer Sciences

Exploring the Virtualization Potential of Raspberry Pi Devices with Proxmox VE

Share

COinS