Sensitivity analysis of scaling factor for a subset of real-time tasks

A hard real-time system is required to complete its operations before all its timing deadlines. For a given task set, it is useful to know what changes to execution time can be made to a subset of tasks that will result in a system that is borderline schedulable. In this paper, we address the sensitivity analysis for scaling factor of a group task execution times for EDF-scheduled systems on a uniprocessor. We prove that an optimal (minimum) scaling factor can be calculated by a single pass of the QPA algorithm, there is no additional search or cost required by the computations. The sensitivity analysis developed for task subset scaling factor are therefore as efficient as QPA, and are easily incorporated into a system design support tool.