No abstract

No abstract

No abstract

This paper proposes a new scheme for global scheduling in a distributed system using a bus network, which is shared by nodes in the system through a contention-free medium access protocol. Unlike existing solutions, our scheme does not require coniplete global state information t,o be collected for a node, which has an unschedulable aperiodic task, to find a node which can execute the task. The nodes are organized in our system in a logical ring fashion to synchronize the global scheduling process. To evaluate the performance of our scheme, two protocols are developed. Simulation results show the performance of the scheme close to an ideal distributed system, in which the centralized global scheduler takes care of aperiodic tasks without any communication and task migration overhead.

This paper presents a distributed computing system, Area-Wide Real-time fiafic Conirol (ARTC) l , designed to provide areawide smooth traffic progress and to prevent frequent occurrences of traffic congestion. The signal controllers in ARTC, which are microcomputers, are interconnected through a computer network. By exchanging traffic flow information among the signal controllers, ARTC provides a new concept in areawide traffic control by exploiting the use of the computer network. Simulation results of the ARTC prototype control algorithm is presented, and the results show substantial improvement over an optimized fixed time control. The signal controllers and the computer network are designed to support the real-time communication requirements and a sufficient level of fault-tolerance.