Journal•ISSN: 0702-0481
International journal of mini & microcomputers
Taylor & Francis
About: International journal of mini & microcomputers is an academic journal. The journal publishes majorly in the area(s): Scheduling (computing) & Fault tolerance. It has an ISSN identifier of 0702-0481. Over the lifetime, 103 publications have been published receiving 444 citations.
Topics: Scheduling (computing), Fault tolerance, Parallel processing (DSP implementation), Parallel algorithm, Petri net
Papers
More filters
Journal Article•
46 citations
Journal Article•
TL;DR: An on-line scheduling approach called TaskPair-Scheduling is presented which merges the concepts of guaranteeing (an activity) and exception handling (due to time-outs) such that the guarantee may enclose the execution of the exception handler.
Abstract: In this paper an online scheduling approach called TaskPairScheduling is presented that merges the concepts of guaranteeing (an activity) and exception handling (due to time-outs) such that the guarantee may enclose the execution of the exception handler. An online scheduling algorithm is shown that plans guaranteed tryexcept constructs as a pair of tasks (Main Task, Except Task) by considering the Except Task aa a hard real-time task (task of maximum importance) and the Main Task by its relative importance (this may depend on arbitary criteria). This leads to timely predictable be havior even if the Main Task of the activity has an arbitrary execution time
34 citations
Journal Article•
29 citations
Journal Article•
TL;DR: In this paper, the authors consider three representative token ring protocols that implement the earliest-deadline first transmission (scheduling) policy to different degrees with different overheads and compare their worst-case performance.
Abstract: In developing distributed real-time scheduling algorithms such as real-time communication protocols, issues in achieving optimal transmission policy and minimizing overhead must be addressed. In this study, we tackle this problem in the context of a specific distributed system: a token ring network. We consider three representative token ring protocols that implement the earliest-deadline first transmission (scheduling) policy to different degrees with different overheads. Their worst-case performance is analyzed and compared. We observe that no protocol can always dominate the others in the entire system parameter space defined by network attributes and application requirements. We conclude that to evaluate a distributed real-time scheduling algorithm such as a communication protocol, one should consider not only the sheduling policy employed but also the overhead incurred in the implementation of the scheduling policy
29 citations
Journal Article•
26 citations