On a laxity-based real-time scheduling policy for fixed-priority tasks and its non-utilization bound

Understanding the end-to-end temporal behavior of distributed systems is a fundamental concern in real-time computing. Meeting timing requirements is critical to many computer applications. These real-time systems employ a schedulability test to determine whether each task can meet its deadline. In this paper, we extend Liu and Abdelzaher´s work by introducing laxity as the priority of the real-time tasks. A Least-Laxity-First (LLF) scheduling algorithm assigns higher priority to a task with the least laxity, and has been proven to be optimal for uniprocessor systems. We compare this laxity-based scheduling policy with deadline monotonic scheduling (DMS), shortest-job-first scheduling (SJF), and velocity monotonic scheduling (VMS). Simulations show that the proposed laxity-based scheduling policy is a feasible approach for real-time admission control.