Towards exploiting the preservation strategy of deferrable servers

Worst-case response time analysis of hard real-time tasks under hierarchical fixed priority pre-emptive scheduling (H-FPPS) has been addressed in a number of papers. Based on an exact schedulability condition, we showed in [4] that the existing analysis can be improved for H-FPPS when deferrable servers are used. In this paper, we reconsider response time analysis and show that improvements are not straightforward, because the worst-case response time of a task is not necessarily assumed for the first job when released at a critical instant. The paper includes a brief investigation of best-case response times and response jitte