Clock Synchronization in Virtualized Distributed Real-Time Systems Using IEEE 802.1AS and ACRN

• Main Authors: Jan Ruh, Wilfried Steiner, Gerhard Fohler
• Format: Article
• Language: English
• Published: IEEE 2021-01-01
• Series: IEEE Access
• Subjects: Hypervisor; clock synchronization; virtualization; ACRN; IEEE 8021AS
• Online Access: https://ieeexplore.ieee.org/document/9530651/

Virtualization of distributed real-time systems enables the consolidation of mixed-criticality functions on a shared hardware platform, easing system integration. Time-triggered communication and computation can act as an enabler of safe hard real-time systems. A <italic>time-triggered hypervisor</italic> that activates virtual CPUs according to a global schedule can provide the means to allow for a resource-efficient implementation of the time-triggered paradigm in virtualized distributed real-time systems. A prerequisite of time-triggered virtualization for hard real-time systems is providing access to a global time base to VMs and the hypervisor. A global time base results from clock synchronization with an upper bound on the <italic>clock synchronization precision</italic>. We present a formalization of the notion of time in virtualized distributed real-time systems. We use this formalization to propose a <italic>virtual clock condition</italic> that enables us to test the suitability of a virtual clock for the design of virtualized time-triggered real-time systems focusing on clock synchronization. We discuss and model how virtualization, particularly resource consolidation versus resource partitioning, degrades clock synchronization precision. Finally, we apply our insights to model the IEEE 802.1AS clock synchronization protocol and derive an upper bound on the clock synchronization precision of IEEE 802.1AS in a virtualized distributed real-time system. We present our implementation of a <italic>dependent clock</italic> for ACRN that can be synchronized to a grandmaster clock. The results of our experiments illustrate that a type-1 hypervisor like ACRN implementing the dependent clock paradigm yields native clock synchronization precision. Furthermore, we show that the upper bound of clock synchronization precision derived from our model holds for a series of experiments featuring native and virtualized setups.