Minimal Virtual Machines on IoT Microcontrollers: The Case of Berkeley Packet Filters with rBPF

TL;DR

This paper evaluates two minimal virtual machine implementations, rBPF and WebAssembly, on low-power IoT microcontrollers, demonstrating that rBPF offers a low-overhead, memory-efficient solution suitable for small, updatable software modules.

Contribution

The paper introduces rBPF, a register-based VM optimized for IoT microcontrollers, and compares its performance and memory overhead with WebAssembly, showing rBPF's suitability for resource-constrained devices.

Findings

rBPF has less than 10% additional memory overhead.

Both VMs have tolerable execution time overhead for low-throughput IoT devices.

rBPF enables small, updatable modules with minimal resource impact.

Abstract

Virtual machines (VM) are widely used to host and isolate software modules. However, extremely small memory and low-energy budgets have so far prevented wide use of VMs on typical microcontroller-based IoT devices. In this paper, we explore the potential of two minimal VM approaches on such low-power hardware. We design rBPF, a register-based VM based on extended Berkeley Packet Filters (eBPF). We compare it with a stack-based VM based on WebAssembly (Wasm) adapted for embedded systems. We implement prototypes of each VM, hosted in the IoT operating system RIOT. We perform measurements on commercial off-the-shelf IoT hardware. Unsurprisingly, we observe that both Wasm and rBPF virtual machines yield execution time and memory overhead, compared to not using a VM. We show however that this execution time overhead is tolerable for low-throughput, low-energy IoT devices. We further show that, while using a VM based on Wasm entails doubling the memory budget for a simple networked IoT application using a 6LoWPAN/CoAP stack, using a VM based on rBPF requires only negligible memory overhead (less than 10% more memory). rBPF is thus a promising approach to host small software modules, isolated from OS software, and updatable on-demand, over low-power networks.