| Hi-index | 0.00 |
In recent years more and more real-time applications run on multiprocessor or distributed systems. In such systems, a task may execute sequentially on many different processors. Such a task can be viewed as a linear chain of subtasks, each of which represents a segment of the task that executes on one of those processors. The response time of the task is measured from the release of its first subtask to the completion of its last subtask and is called the end-to-end response time. A task is schedulable if its end-to-end response time is never greater than the specified end-to-end relative deadline. This thesis deals with the problem of scheduling periodic tasks to meet their end-to-end deadlines. Specifically, the thesis focuses on fixed-priority scheduling algorithms, where each subtask is assigned a fixed priority and is scheduled preemptively. According to this approach, three related problems need to be solved. Priority Assignment: How we assign the priorities to subtasks so that the system schedulability can be maximized. Execution Synchronization: How we synchronize the releases of subtasks so that the precedence constraints among subtasks are satisfied, and the system schedulability is optimized. Schedulability Analysis: Given a certain priority assignment and an execution synchronization method, how we compute the worst-case end-to-end response time of each task, so that we can verify the schedulability of the task and hence the schedulability of the system. As solutions to these three problems, the thesis describes five synchronization protocols, several deadline-based priority assignment methods, and corresponding schedulability analysis algorithms. These solutions form an integrated end-to-end scheduling framework for scheduling tasks with end-to-end deadlines and analyzing the schedulability of an end-to-end system. As an application of the framework, the thesis proposes an end-to-end scheduling based approach to the resource contention problem in distributed real-time systems. Simulation results show that the end-to-end scheduling approach performs better in many cases than an existing approach, known as the Multiprocessor Priority Ceiling Protocol. In the appendix, the thesis includes an in-depth discussion of sporadic server algorithms.