The Adaptive Scheduling Approaches to Open Real-Time Systems
Yong Zou · Chinese Journal of Computers · 2004
The adaptive scheduling problems for open real time systems are discussed in this paper. In an open real time system, tasks are loaded dynamically and their schedulability is tested on line. At the mean time, how much CPU bandwidth the tasks need may not be known as a priori. This feature urges us to develop adaptive real time scheduling approaches which can allocate resources to tasks dynamically at run time. According to the scheduling requirements of the open real time system, adaptive scheduling approaches are presented respectively for hard real time tasks and soft real time tasks. Firstly, an approach based on feedback control is presented for the hard real time tasks scheduling. Secondly, for soft real time applications, another adaptive scheduling approach is presented. The second one is based on fussy control technology. It can control the deadline missing ratio near the expected value while sampling the applications’ performance. These adaptive scheduling approaches are developed for open real time system, with the merits that they can schedule different kinds of real time tasks and adapt to their dynamic resource requirements through tuning the schedulers.