Even today, safety-critical systems in many fields of application use separate processors to isolate software of different criticality from another. The resulting system architecture is non-optimal in regard to flexibility, device size and power consumption. These drawbacks can be prevented by the use of partitioning operating systems that enable the integration of applications with different criticality on a single processor. However, their application for deeply-embedded devices, that are characterized by strict resource constraints and the lack of advanced processor features such as memorymanagement units (MMU), is challenging. In this work, we show that the impact of virtualization on performance and predictability is smaller in the field of deeply-embedded devices than in more complex systems, making it a compelling choice as a partitioning technology. We present a hypervisor that provides time and space partitioning for an MMU-less system, as well as mechanisms for communication and resource sharing. To satisfy the strict power and resource constraints found in deeply-embedded devices, we focus on solutions with a minimal runtime overhead. Furthermore, the hypervisor is integrated with the processor power management, often enabling significant power savings in the resulting system architecture.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.