Partitioned Scheduling Algorithm Research Articles

Real-Time Scheduling for Periodic Tasks on Uniform Multiprocessors

The problem of scheduling a set of periodic tasks on a uniform multiprocessor system is considered in the present study. Each processor in a uniform multiprocessor system is characterized by its speed or computation capacity, i.e., execution of a job on a processor with speed s for t time units completes s × t units of execution. In the commonly-known partitioned scheduling, each task is assigned to a processor and all of its jobs are required to be executed on that processor. However, partitioning of periodic tasks requires solving the bin-packing problem, which is known to be intractable (NPhard in the strong sense). This paper presents a global scheduling algorithm that transforms a given periodic task system into another using a “task-splitting” (as opposed to the “set-splitting”) technique. Each transformed periodic task system is guaranteed to be scheduled successfully on any uniform multiprocessor using a partitioned scheduling algorithm. The earliest deadline first (EDF) algorithm is chosen for scheduling tasks on each processor. It is proven that the proposed algorithm results in the theoretical-maximum utilization bound on any uniform multiprocessor platform if the platform is “reasonably powerful”. Therefore, the proposed algorithm is optimal in the sense of maximizing achievable utilization. Since the task-splitting technique will incur context switches during runtime, we have also considered the ways of reducing the number of context switches, and suggest a method which can significantly reduce the number of context switches in the schedules generated by the proposed algorithm.

Pessimism in multicore global schedulability analysis

Abstract When it comes to multicore real-time systems, the global scheduling that allows tasks to migrate between different cores and the partitioned scheduling that statically allocates tasks to individual cores are two popular policies. However, due to the complexity and pessimism of multicore schedulability tests, there is no simple way to compare the effectiveness of global versus partitioned scheduling, and it is hard to improve existing analysis methods. In this paper, we formally prove that, according to state-of-the-art global schedulability tests, if a multicore task system (with Di ≤ Ti) is deemed schedulable under global fixed priority policy, we can easily find a schedulable partitioning for it too. This result holds also for self-suspending tasks that can suspend its own execution voluntarily. Such a fact reveals the high pessimism in current global schedulability tests. To tackle this, we propose a 2-part execution scenario, by simply considering the task execution into two parts, that helps better estimate the interference a task suffers. Using this method, a new global schedulability test can detect schedulable tasksets that will be non-schedulable by any partitioned scheduling algorithm. Our proposed method indicates a new direction for designing global schedulability tests. In the experiments, comparisons with other global and partitioned schedulability tests confirm the improvement by using our new test.

Resource Optimization for Real-Time Streaming Applications Using Task Replication

In this paper, we study the problem of exploiting parallelism in a hard real-time streaming application modeled as an acyclic synchronous data flow (SDF) graph and scheduled on a heterogeneous multiprocessor system-on-chip platform to alleviate the capacity fragmentation due to partitioned scheduling algorithms and reduce the number of required processors when a throughput requirement is satisfied. As the main contribution in this paper, we propose a method to determine a replication factor for each task in an acyclic SDF graph such that by distributing the workloads among more parallel tasks with lower utilization in the obtained transformed graph, the left capacity on the processors can be efficiently exploited, hence reducing the number of required processors. The experimental results, on a set of real-life streaming applications, demonstrate that our approach can reduce the minimum number of processors required to schedule an application and considerably improve the memory requirements and application latency compared to related approaches while meeting the same throughput constraint.

Enhanced fixed-priority real-time scheduling on multi-core platforms by exploiting task period relationship

One common approach for multi-core partitioned scheduling problem is to transform this problem into a traditional bin-packing problem, with the utilization of a task being the “size” of the object and the utilization bound of a processing core being the “capacity” of the bin. However, this approach ignores the fact that some implicit relations among tasks may significantly affect the feasibility of the tasks allocated to each local core. In this paper, we study the problem of partitioned scheduling of periodic real-time tasks on multi-core platforms under the Rate Monotonic Scheduling (RMS) policy. We present two effective and efficient partitioned scheduling algorithms, i.e. PSER and HAPS, by exploiting the fact that the utilization bound of a task set increases as task periods are closer to harmonic on a single-core platform. We formally prove the schedulability of our partitioned scheduling algorithms. Our extensive experimental results demonstrate that the proposed algorithms can significantly improve the scheduling performance compared with the existing work.

Partitioned scheduling of parallel real-time tasks on multiprocessor systems

In this paper, we focus on the scheduling of periodic fork-join real-time tasks on multiprocessor systems. Parallel real-time tasks in the fork-join model have strict parallel segments without laxity. We propose a partitioned scheduling algorithm which increases the laxity of the parallel segments and therefore the schedulability of tasksets of this model. A similar algorithm has been proposed in the literature but it produces job migrations. Our algorithm eliminates the use of job migrations in order to create a portable algorithm that can be implemented on a standard Linux kernel. Results of extensive simulations are provided in order to analyze the schedulability of the proposed algorithm, and to provide comparisons with the other algorithm proposed in the literature.

Job vs. portioned partitioning for the earliest deadline first semi-partitioned scheduling

In this paper, we focus on the semi-partitioned scheduling of sporadic tasks with constrained deadlines and identical processors. We study two cases of semi-partitioning: (i) the case where the worst case execution time (WCET) of a job can be portioned, each portion being executed on a dedicated processor, according to a static pattern of migration; (ii) the case where the jobs of a task are released on a processor, 1 time out of p, where p is an integer at most equal to the number of processors, according to a round-robin migration pattern. The first approach has been investigated in the state-of-the-art by migrating a job at its local deadline, computed from the deadline of the task it belongs to. We study several local deadline assignment heuristics (fair, based on processor utilization and based on the minimum acceptable local deadline for a job on a processor). In both cases, we propose feasibility conditions for the schedulability of sporadic tasks scheduled using earliest deadline first (EDF) semi-partitioned scheduling. We show that the load function used for global scheduling to establish the feasibility of sporadic task sets exhibits interesting properties in the semi-partitioning context. We carry out simulations to study the performance of the two approaches in terms of success rate and number of migrations, for platforms composed of four and eight processors. We compare the performance of these semi-partitioned heuristics with the performance of classical partitioned scheduling algorithms and with a global scheduling heuristic which is currently considered to have good performances.

유니폼 멀티프로세서 환경에서 단순 주기성 태스크를 위한 최적 RM 스케줄링

본 논문에서는 유니폼 멀티프로세서 환경에서 단순 주기성 태스크 시스템을 성공적으로 스케줄 할 수 있는 알고리즘을 제안한다. 멀티프로세서 환경에서 주기성 태스크를 스케줄하기 위한 파티션드(partitioned) 스케줄링 알고리즘은 bin-packing 문제와 같은 문제로써 해결하는 게 불가능하다고 알려져 있다. 본 논문에서는 "task-splitting"기법을 이용하여 단순 주기성 태스크 시스템을 다른 단순 주기성 태스크 시스템으로 변환하는 글로벌(global) 스케줄링 알고리즘을 제시하고, 변환과정을 거친 단순 주기성 태스크 시스템은 유니폼 멀티프로세서에서 파티션드 스케줄링 알고리즘에 의해 성공적으로 스케줄 된다. 그리고 유니폼 멀티프로세서 환경에서 제안한 알고리즘이 이론적으로 최대 이용률 범위(utilization bound)까지 성공적으로 스케줄 할 수 있음을 증명한다. The problem of scheduling simply periodic task systems upon a uniform multiprocessor is considered. Partitioning of periodic task systems requires solving the bin-packing problem, which is known to be intractable (NP-hard in the strong sense). This paper presents a global scheduling algorithm which transforms a given simply periodic task system into another using a "task-splitting" technique. Each transformed simply periodic task system is guaranteed to be successfully scheduled upon any uniform multiprocessor using a partitioned scheduling algorithm. It is proven that the proposed algorithm achieves the theoretical maximum utilization bound upon any uniform multiprocessor platform.