Guaranteeing Pfair supertasks by reweighting

Holman, Philip; Anderson, James H.

doi:10.1109/real.2001.990612

Cited by 18 publications

(37 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, it has been proved that Pfair schedules with such mapping constraints do exist [19]. Based on the concept of supertask, a reweighting technique has been studied, which inflates a supertask's weight to ensure that its component tasks meet their deadlines if the supertask is scheduled in a Pfair manner [14]. While this technique ensures that the supertask's non-migratory component tasks meet their deadlines, some system utilization is sacrificed.…”

Section: Related Workmentioning

confidence: 99%

An optimal boundary fair scheduling algorithm for multiprocessor real-time systems

Zhu

Mossé

et al. 2011

Journal of Parallel and Distributed Computing

View full text Add to dashboard Cite

Although the scheduling problem for multiprocessor real-time systems has been studied for decades, it is still an evolving research field with many open problems. In this work, focusing on periodic real-time tasks, we propose a novel optimal scheduling algorithm, namely boundary fair (Bfair), which follows the same line of research as the well-known Pfair scheduling algorithms and can also achieve full system utilization. However, different from the Pfair algorithms that make scheduling decisions at every time unit to enforce proportional progress (i.e., fairness) for each task, Bfair makes scheduling decisions and enforces fairness to tasks only at tasks' period boundaries. The correctness of the Bfair algorithm to meet the deadlines of all tasks' instances * A preliminary version of this paper appeared in IEEE RTSS 2003. This work is supported in part by NSF award CNS-0720651. 1 is formally proved. The performance of Bfair is evaluated through extensive simulations. The results show that, compared to that of the Pfair algorithms, Bfair can significantly reduces the number of scheduling points (by upto 94%) and the time overhead of Bfair is comparable to that of the most efficient Pfair algorithm (i.e., P D 2 ). Moreover, by aggregating the time allocation of tasks for the time interval between consecutive period boundaries, the resulting Bfair schedule needs dramatically reduced number of context switches and task migrations, as low as 18% and 15%, respectively, when compared to those of Pfair schedules.

show abstract

Section: Related Workmentioning

confidence: 99%

An optimal boundary fair scheduling algorithm for multiprocessor real-time systems

Zhu

Mossé

et al. 2011

Journal of Parallel and Distributed Computing

View full text Add to dashboard Cite

show abstract

“…In addition, unlike a supertask, the fractional part (f ) of a megatask's ideal weight may be less than W max . Hence, there is not much semblance between the approach used in this paper and that in [13].…”

Section: Megatasksmentioning

confidence: 93%

“…For this, first consider three tasks with weights On the other hand, if we were to use supertasking, then at least two supertasks will be required. Even with the better of the two possible packings, wherein the first supertask contains the first two component tasks, and the second contains just the third task, using the rules of [13], the weight of the first supertask will have to be increased by 7 24 , which is 1 4 higher than 1 24 . Thus, in this case supertasking entails more overhead.…”

Section: Megatasksmentioning

confidence: 99%

Real-Time Scheduling on Multicore Platforms

Anderson

Calandrino

Devi

12th IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS'06)

Self Cite

View full text Add to dashboard Cite

Multicore architectures, which have multiple processing units on a single chip, are widely viewed as a way to achieve higher processor performance, given that thermal and power problems impose limits on the performance of single-core designs. Accordingly, several chip manufacturers have already released, or will soon release, chips with dual cores, and it is predicted that chips with up to 32 cores will be available within a decade. To effectively use the available processing resources on multicore platforms, software designs should avoid co-executing applications or threads that can worsen the performance of shared caches, if not thrash them.While cache-aware scheduling techniques for such platforms have been proposed for throughput-oriented applications, to the best of our knowledge, no such work has targeted real-time applications. In this paper, we propose and evaluate a cache-aware Pfair-based scheduling scheme for real-time tasks on multicore platforms.

show abstract

“…Unfortunately, the supertask approach cannot ensure all the non-migratory component tasks to meet their deadline even when the supertask is scheduled in a Pfair manner. 1 Based on the concept of supertask, a re-weighting technique has been studied, which inflates a supertask's weight to ensure that its component tasks meet their deadlines if the supertask is scheduled in a Pfair manner (Holman and Anderson 2001). While this technique ensures that the supertask's non-migratory component tasks meet their deadlines, the available system utilization is reduced.…”

Section: Related Workmentioning

confidence: 99%

Cluster scheduling for real-time systems: utilization bounds and run-time overhead

Zhu

Aydın

2011

Real-Time Syst

View full text Add to dashboard Cite

Cluster scheduling, where processors are grouped into clusters and the tasks that are allocated to one cluster are scheduled by a global scheduler, has attracted attention in multiprocessor real-time systems research recently. In this paper, assuming that an optimal global scheduler is adopted within each cluster, we investigate the worst-case utilization bounds for cluster scheduling with different task allocation/partitioning heuristics. First, we develop a lower limit on the utilization bounds for cluster scheduling with any reasonable task allocation scheme. Then, the lower limit is shown to be the exact utilization bound for cluster scheduling with the worstfit task allocation scheme. For other task allocation heuristics (such as first-fit, bestfit, first-fit decreasing, best-fit decreasing and worst-fit decreasing), higher utilization bounds are derived for systems with both homogeneous clusters (where each cluster has the same number of processors) and heterogeneous clusters (where clusters have different number of processors). In addition, focusing on an efficient optimal global scheduler, namely the boundary-fair (Bfair) algorithm, we propose a period-aware task allocation heuristic with the goal of reducing the scheduling overhead (e.g., the number of scheduling points, context switches and task migrations). Simulation results indicate that the percentage of task sets that can be scheduled is significantly improved under cluster scheduling even for small-size clusters, compared to that of the partitioned scheduling. Moreover, when comparing to the simple generic task Real-Time Syst (2011) 47: 253-284 allocation scheme (e.g., first-fit), the proposed period-aware task allocation heuristic markedly reduces the scheduling overhead of cluster scheduling with the Bfair scheduler.

show abstract

Guaranteeing Pfair supertasks by reweighting

Cited by 18 publications

References 7 publications

An optimal boundary fair scheduling algorithm for multiprocessor real-time systems

An optimal boundary fair scheduling algorithm for multiprocessor real-time systems

Real-Time Scheduling on Multicore Platforms

Cluster scheduling for real-time systems: utilization bounds and run-time overhead

Contact Info

Product

Resources

About