Real-time System Overheads: a Literature Overview

In most contemporary systems there are several jobs concurrently competing for shared resources, such as a processor, memory, network, sensors or other devices. Sharing a resource between several jobs requires synchronizing the jobs, specifying when which job will have access to the resource. A common synchronization method is scheduling. Executing a schedule requires switching resource assignments between the jobs, which is usually referred to as context switching. The overheads associated with scheduling and context switching are part of the system overheads. Initially, in the spirit of keeping things simple, real-time systems analysis abstracted from many details, including the overheads incurred by the operating system. This has led to inherently optimistic results, i.e. accepting collections of jobs, which if executed on a real system will fail to meet all the constraints. In this paper we consider a less idealized platform by taking some practical aspects into account. We present an overview of literature dealing with real-time system overheads, in particular the scheduling and context switch overheads. We focus on sharing a single timely resource, such as a processor, in the context of Fixed Priority Preemptive Scheduling. We treat in detail the overheads due to preemption, and due to blocking of tasks for different resource access protocols (which is the main contribution of this paper).