A model-based approach for multiple QoS in scheduling: from models to implementation
Automated Software Engineering
| Hi-index | 0.00 |
Fixed priority scheduling (FPS) has been widely studiedand used in a number of applications, mostly due to itsflexibility, simple run-time mechanism and small overhead.However, preemption related overhead in FPS may causeundesired high processor utilization, high energy consumption,or, in some cases, even infeasibility.In this paper, we propose a method to reduce the numberof preemptions in legacy FPS systems consisting oftasks with priorities, periods and offsets. Unlike other approaches,our algorithm does not require modification of thebasic FPS mechanism. Our method analyzes off-line a setof periodic tasks scheduled by FPS, detects the maximumnumber of preemptions that can occur at run-time, and reassignstask attributes such that the tasks are schedulableby the same scheduling mechanism, while achieving a significantlylower number of preemptions.In some cases, there is a cost to pay for a lower numberof preemptions in terms of increased amount of tasks and/orreduced task execution flexibility. Our method provides forthe ability to choose a user-defined number of preemptionswith respect to the cost to pay.