An anomaly prevention approach for real-time task scheduling
Journal of Systems and Software
| Hi-index | 0.00 |
Many time-critical applications require predictableperformance and tasks in these applications have deadlines to be met despite the presence of faults. In this paper, we propose a new dynamic non-preemptive scheduling algorithm for a relatively new task model called parallelizable task model where real-time tasks can be executed concurrently on multiple processors. We usethis parallelism in tasks to meet their deadlines andthus obtain better processor utilization compared to non-parallelizable task scheduling algorithms. We assumethat tasks are aperiodic. Further, each task is characterized by its deadline, resource requirements, and worstcase computation time on p processors, where p is thedegree of task parallelization. To study the effectiveness of our algorithm, we have conducted extensive simulation studies and compared its performance with themyopic1scheduling algorithm [8]. We found that thesuccess ratio offered by our algorithm is always higherthan the myopic algorithm for a wide variety of task parameters. Also, we propose a resource reclaiming algorithm to reclaim resources from parallelizable real-timetasks when their actual computation times are less thantheir worst case computation times. Our parallelizabletask scheduling together with its associated reclaimingoffers the best guarantee ratio compared to the other algorithmic combinations.