I. Introduction
Platform virtualization, which enables multiple operating systems (OSes) to run on top of the same hardware platform, is gaining momentum in the embedded systems domain [1]. Nevertheless, embedded virtualization has inherent characteristics which substantially differentiate it from general-purpose or server virtualization, since embedded devices are typically resource and power constrained systems that must respond to events within strict deadlines [2]. While in the server and desktop space virtualization is used for load balancing, service consolidation and power management, in the embedded systems domain virtualization has been primarily employed in order to partition functionality (real-time and non-real-time characteristics), as well as minimizing the design and certification efforts [2], [3].