Research on Xen Virtualization and Scheduling Algorithm Based on Mixed-Criticality Systems

This paper analyzes the real-time guarantee problem and CPU resource allocation problem of Xen scheduling in mixed criticality systems, and proposes a virtual machine scheduling idea for Xen mixed-criticality systems. By adjusting VCPU resources under different load conditions, the CPU resources occupied by non-real-time virtual machines are reasonably allocated while ensuring real-time performance. In view of the passive treatment of low-criticality tasks in mixed criticality systems, a preemptive mixed criticality task global scheduling method SWPA is proposed. The algorithm is based on a dynamic weight adjustment factor, combined with the criticality and time sensitivity of the task, dynamically evaluates the importance of the task, and assigns priority to the task. The correctness of the method is proved by theoretical deduction. Simulation experiments show that this method is superior to the existing classic scheduling algorithms EDF-DVD and Credit in terms of the schedulability of low-criticality tasks.