@article{M94B146D9,
title = "Analysis of Workload Performance and Resource Consumption of Lightweight Secure Virtualization Platforms in IoT/Edge Computing",
journal = "The Journal of Korean Institute of Communications and Information Sciences",
year = "2024",
issn = "1226-4717",
doi = "10.7840/kics.2024.49.10.1386",
author = "Eu Jin Ko, Wonmi Choi, Gyeongsik Yang, Chuck Yoo",
keywords = "IoT, Edge Computing, Secure Virtualization Platforms, Network Performance Analysis",
abstract = "Recently, lightweight and secure virtualization platforms that combine the high security of virtual machines with the lightweight nature (high performance) of containers have been widely investigated. These platforms are considered crucial for IoT devices and gateways, which create and communicate vast amounts of data with a limited amount of computing resources. Although various platforms have been proposed, to our knowledge, no studies have compared these platforms in IoT devices and edge computing environments in terms of understanding practical workload performance and CPU usage. Consequently, it is challenging to determine a proper platform to achieve optimal performance and efficient resource consumption. To this end, this study compares and analyzes the latest lightweight secure virtualization platforms―Firecracker, Kata container, and Unikraft―in edge computing environments. Using the Redis program, which is frequently used for data collection and storage, the study details the analysis of each platform’s requests per second and CPU usage.
The results show that the proper platform varies based on the message size being transmitted in edge computing, and the choice of platform affects the CPU resources available for user applications. We believe this study provides a baseline for selecting the appropriate platform in actual edge computing environments and serves as a basis for improving future lightweight, secure virtualization platforms."
}