In recent years, extensive research has been carried out on real-time typed scheduling and analysis of parallel tasks represented as directed acyclic graphs (DAGs) executed on heterogeneous multicores. Although previous studies have examined the schedulability of typed DAG tasks, they still encounter pessimism caused by interference from other tasks. In this article, we explore the worst case response time (WCRT) analysis of typed scheduling for DAG tasks under global scheduling. Here, each vertex in a typed DAG task experiences interference from within itself as well as from higher priority tasks. First, we propose an efficient method to bound the WCRT of typed DAG tasks based on the state-of-the-art parallel task analysis approach. Then, we discover a technique to mitigate the pessimism caused by other tasks, albeit in a nonoptimal manner. Finally, we conduct experiments using randomly generated typed DAG tasks to evaluate the performance of our proposed methods. The results indicate that our proposed approach can yield less pessimistic WCRT under global scheduling.

In recent years, the maturation of modern multicore processor technology and its increasing adoption in critical industrial domains have posed significant challenges for real-time systems, primarily due to contention for shared cache resources and the resulting uncertainty. To address this issue, contemporary processors employ cache partitioning techniques, enhancing temporal predictability by isolating cache access among processor cores. However, this isolation technique may lead to real-time tasks missing their deadlines due to an insufficient number of cache partitions. Consequently, this paper investigates the schedulability of preemptive global Earliest Deadline First (EDF) real-time scheduling algorithms that support dynamic cache allocation. We propose an innovative LAG-based schedulability analysis method for these algorithms and present a utilization-based schedulability condition that reduces analysis time complexity while improving analysis accuracy. Lastly, the performance and efficiency of the proposed schedulability determination method are validated through simulation experiments with randomly generated tasks.

Heterogenerous multi-cores utilize the strength of different architectures for executing particular types of workload, and usually offer higher performance and energy efficiency. In this paper, we study the worst-case response time (WCRT) analysis of typed scheduling of parallel DAG tasks on heterogeneous multi-cores, where the workload of each vertex in the DAG is only allowed to execute on a particular type of cores. The only known WCRT bound for this problem is grossly pessimistic and suffers the non-self-sustainability problem. In this paper, we propose two new WCRT bounds. The first new bound has the same time complexity as the existing bound, but is more precise and solves its non-self-sustainability problem. The second new bound explores more detailed task graph structure information to greatly improve the precision, but is computationally more expensive. We prove that the problem of computing the second bound is strongly NP-hard if the number of types in the system is a variable, and develop an efficient algorithm which has polynomial time complexity if the number of types is a constant. Experiments with randomly generated workload show that our proposed new methods are more precise than the existing bound while having good scalability.

Heterogenerous multi-cores utilize the strength of different architectures for executing particular types of workload, and usually offer higher performance and energy efficiency. In this paper, we study the worst-case response time (WCRT) analysis of typed scheduling of parallel DAG tasks on heterogeneous multi-cores, where the workload of each vertex in the DAG is only allowed to execute on a particular type of cores. The only known WCRT bound for this problem is grossly pessimistic and suffers the non-self-sustainability problem. In this paper, we propose two new WCRT bounds. The first new bound has the same time complexity as the existing bound, but is more precise and solves its non-self-sustainability problem. The second new bound explores more detailed task graph structure information to greatly improve the precision, but is computationally more expensive. We prove that the problem of computing the second bound is strongly NP-hard if the number of types in the system is a variable, and develop an efficient algorithm which has polynomial time complexity if the number of types is a constant. Experiments with randomly generated workload show that our proposed new methods are significantly more precise than the existing bound while having good scalability.

The analysis of global multiprocessor scheduling is more difficult than its uniprocessor counterpart. Due to the unknown critical instant, existing techniques use over-approximations of task interference for efficient yet pessimistic analysis. In this paper, we proposed a new technique to improve the precision of interference estimation. The key is to identify and resolve contradicting assumptions made in the analysis procedure. The resulting new analysis method improves the analysis precision at the price of a higher complexity. Then we introduce techniques to optimize the new method for better efficiency. Experiments with randomly generated task sets are conducted to evaluate both the precision and efficiency of the proposed new method.