A new technique for analyzing soft real-time self-suspending task systems
Cong Liu, James H. Anderson · ACM SIGBED Review · 2012
We consider the problem of globally scheduling soft real-time sporadic self-suspending task systems on multiprocessors. Existing analysis methods are pessimistic, yielding O(n) utilization loss where n is the number of tasks in the system. Unless the number of tasks is small and suspension delays are short, such methods entail significant capacity loss. We identify the fundamental sources that cause pessimism in existing methods, and propose a new analysis technique that entails only O(m) suspension-related utilization loss, where m is the number of processors.