Performance Comparison of ESP-r in Containers vs Virtual Machines for Digital Twins

This study compares the performance of the ESP-r building energy modeling tool in containerized environments (Docker, LXD, LXC) versus traditional virtual machines (VMs, VirtualBox, VMware), focusing on execution time, resource utilization, and scalability. Given the increasing adoption of cloud computing and virtualization, it's crucial to identify the most efficient platform for ESP-r simulations. The hypothesis suggests that containers, due to their lightweight architecture, may offer superior performance compared to VMs. The research involves setting up ESP-r in both environments, conducting simulations, and recording execution times to determine the optimal virtualization technology. Additionally, the study examines the role of these technologies in supporting digital twins, dynamic virtual representations of physical systems used for performance monitoring and decision-making in the construction industry.

I. Introduction

DIGITAL twins, as dynamic virtual representations of physical systems, have become increasingly significant in the realm of building performance simulation. They serve as a bridge between the physical and digital worlds, enabling real-time monitoring, analysis, and decision-making. In the context of urbanization and technological advancement, the demand for efficient energy systems has intensified, and digital twins have emerged as a transformative solution in this domain.