Engineering and analysis of fixed priority schedulers

Abstract

Scheduling theory holds great promise as a means to a priori validate timing correctness of real-time applications. However, there currently exists a wide gap between scheduling theory and its implementation in operating system kernels running on specic hardware platforms. The implementation of any particular scheduling algorithm introduces overhead and blocking components which must be accounted for in the timing correctness validation process. This paper presents a methodology for incorporating the costs of scheduler implementation within the context of xed priority scheduling algorithms. Both event-driven and timerdriven scheduling implementations are analyzed. We show that for the timer-driven scheduling implementations the selection of the timer interrupt rate can dramatically aect the schedulability of a task set, and we present a method for determining the optimal timer rate. We analyzed both randomly generated and two well dened task sets and found that their schedulability can be signicantly degraded by the implementation costs. Task sets that have ideal breakdown utilization over 90% may not even be schedulable when the implementation costs are considered. This work provides a rst step towards bridging the gap between realtime scheduling theory and implementation realities. This gap must be bridged for any meaningful validation of timing correctness properties of real-time applications. Keywords| Real-time, scheduling, periodic, schedulability, feasibility. This research is supported in part by grants from the Oce of Naval Research and the Naval Ocean Systems Center under contract N00014-91-J-1304 scheduling theory and its implementation in operating system kernels running on specic hardware platforms. This work provides a rst step towards bridging the gap between real-time scheduling theory and implementation realities. This gap must be bridged for any meaningful validation of timing correctness properties of real-time applications. This paper will take into account the costs of the kernel scheduling mechanisms, which in turn are a function of the underlying hardware support. We dene the kernel costs as either overhead or blocking. Overhead is the time spent in the kernel performing a service on behalf of a specic task, such as invoking or terminating it. Blocking, or priority inversion, is time spent, either

Response-time analysis for fixed-priority systems with a write-back cache

“…Early work on accounting for scheduling overheads in fixed-priority preemptive systems by Katcher et al (1993) and Burns (1994) focused on scheduler overheads and context switch costs. Subsequent work on the analysis of Cache Related Preemption Delays (CRPD) and their integration into schedulability analyses used the concepts of Useful Cache Blocks (UCBs) and Evicting Cache Blocks (ECBs), see Sect.…”

Power optimization of real-time embedded systems on variable speed processors

“…First, f max and V max are obtained for each application using (7) and (9), which are summarized in Table 2. Clearly, they are smaller with EDF than with FPS, because EDF sets the lower bound for f max and V max .…”

“…To exploit these idle intervals, we rely on a powerefficient version of priority-based real-time scheduing method [4], which we call ÐÔÔ× for brevity. It is based on the implementation model of priority-based scheduling in the kernel [9]. The priority-based scheduling can be implemented by maintaining two queues, one called run queue and the other called delay queue.…”

Extensiones Al Lenguaje Ada Y a Los Servicios Posix Para Planificación en Sistemas De Tiempo Real Estricto.

“…Según la clasificación de Katcher [35], MaRTE OS es un sistema dirigido por eventos no integrado, es decir, el temporizador hardware se programa para que interrumpa en el siguiente instante de tiempo relevante, y esta acción puede interrumpir la ejecución de cualquier tarea, independientemente de su prioridad. En este tipo de sistemas, la sobrecarga del sistema en una aplicación de tiempo real se produce de cuatro formas distintas:…”