论文部分内容阅读
随着计算机应用的发展，实时任务在控制中应用的范围和规模也越来越大。对于强实时任务，必须保证它的急迫的时间要求。使用多处理器系统来构造实时系统具有较高的性能价格比和良好的扩充性；并能够简化应用的实现。对于周期性任务和随机任务共存的多处理器系统，选择一个最佳的任务管理策略是ＮＰ问题。在本文中，讲述了一般的任务管理策略，并给出了一种基于资源保留策略的集中分配的多处理器实时任务管理策略。
With the development of computer applications, real-time tasks in the control of the scope and scale is also growing. For strong real-time tasks, it must ensure that its urgent time requirements. The use of multiprocessor systems to construct real-time systems offers high cost performance and scalability; and simplifies application implementation. For multiprocessor systems with periodic tasks and random tasks coexisting, choosing an optimal task management strategy is an NP problem. In this paper, we describe the general task management strategy and give a centralized strategy of multi-processor real-time task management based on resource reservation strategy.