Virtualization & Microservice Architecture for Software-Defined Vehicles: An Evaluation and Exploration
Long Wen, Markus Rickert, Fengjunjie Pan, Jianjie Lin, Yu Zhang, Tobias Betz, Alois Knoll
TL;DR
This work evaluates virtualization and containerization for software-defined vehicles across embedded, AMD64, and ARM64 platforms, assessing CPU, memory, network, and disk performance, and validating with Autoware-based automotive workloads. It demonstrates near bare-metal performance for CPU, memory, and networking under containers and lightweight VMs, while disk I/O experiences more pronounced penalties, especially under KVM. The study further proposes a microservice-based architecture to decompose Autoware into modular containers, achieving up to 18% faster startup on AMD64 in function-level configurations and enabling more flexible maintenance. Across hardware platforms, the results show virtualization and containerization are feasible for SDVs, with disk choice and container orchestration significantly impacting startup times and IO patterns. Overall, the paper provides actionable benchmarks and architectural guidance for deploying SDV software using virtualization, containers, and microservices in real-world automotive environments.
Abstract
The emergence of Software-Defined Vehicles (SDVs) signifies a shift from a distributed network of electronic control units (ECUs) to a centralized computing architecture within the vehicle's electrical and electronic systems. This transition addresses the growing complexity and demand for enhanced functionality in traditional E/E architectures, with containerization and virtualization streamlining software development and updates within the SDV framework. While widely used in cloud computing, their performance and suitability for intelligent vehicles have yet to be thoroughly evaluated. In this work, we conduct a comprehensive performance evaluation of containerization and virtualization on embedded and high-performance AMD64 and ARM64 systems, focusing on CPU, memory, network, and disk metrics. In addition, we assess their impact on real-world automotive applications using the Autoware framework and further integrate a microservice-based architecture to evaluate its start-up time and resource consumption. Our extensive experiments reveal a slight 0-5% performance decline in CPU, memory, and network usage for both containerization and virtualization compared to bare-metal setups, with more significant reductions in disk operations-5-15% for containerized environments and up to 35% for virtualized setups. Despite these declines, experiments with actual vehicle applications demonstrate minimal impact on the Autoware framework, and in some cases, a microservice architecture integration improves start-up time by up to 18%.