Towards compositionality in real-time resource partitioning based on regularity bounds

In real-time resource partitioning, a shared resource is partitioned by a resource-level scheduler such that each partition is accessible only by an individual application task group. Tasks within the same task group are scheduled by an application-task-level scheduler that is specialized to the real-time requirements of the tasks in the group. An ideal goal for resource partitioning in real-time systems is to achieve a complete separation of concerns so that: (1) each task group may be executed as if it had access to its own dedicated resource, and (2) there is minimal interaction between the resource-level scheduler and the application-task-level scheduler. In [15], we introduced the notion of a real-time virtual resource which operates at a fraction of the rate of the shared physical resource and whose rate of operation varies with time but is bounded. In this paper we discuss an approach to bound the variation of the rate of operation of a real-time virtual resource by characterizing the rate variation from both temporal and supply dimensions and by expanding on the concept of regularity that was first introduced in [19]. For the case of regular resource partitioning, we show that the utilization bounds of both fixed-priority scheduling and dynamic-priority scheduling remain unchanged from those for dedicated resources. We determine the utilization bounds for the more general case of irregular partitioning. In particular, both types of partitions can be efficiently constructed by exploiting compositionality, properties vis-a-vis the regularity measure.