Research Note
On-line scheduling of scalable real-time tasks on multiprocessor systems

https://doi.org/10.1016/j.jpdc.2003.06.002Get rights and content

Abstract

The computation time of scalable tasks depends on the number of processors allocated to them in multiprocessor systems. As more processors are allocated to a scalable task, the overall computation time of the task decreases but the total amount of processors’ time devoted to the execution of the task, called workload, increases due to parallel execution overhead. In this paper, we propose a task scheduling algorithm that utilizes the property of scalable tasks for on-line and real-time scheduling. In the proposed algorithm, the total workload of all scheduled tasks is reduced by managing processors allocated to the tasks as few as possible without missing their deadlines. As a result, the processors in the system have less load to execute the scheduled tasks and can execute more newly arriving tasks before their deadlines. Simulation results show that the proposed algorithm performs significantly better than the conventional algorithm based on a fixed number of processors to execute each task.

References (24)

  • D. Babbar, P. Krueger, On-line hard real-time scheduling of parallel tasks on partitionable multiprocessors, Internat....
  • E. Bampis, A. Kononov, On the approximability of scheduling multiprocessor tasks with time-dependent processor and time...
  • S. Baruah et al.

    On-line scheduling to maximize task completions

    Proc. IEEE Real-Time Systems Symp.

    (1994)
  • K. Belkhale et al.

    Approximation algorithm for the partitionable independent task scheduling problem

    Internat. Conf. Parallel Process.

    (1990)
  • A. Burchard, J. Liebeherr, Y. Oh, S. Son, New strategies for assigning real-time tasks to multiprocessor systems, IEEE...
  • S. Chiang et al.

    Use of application characteristics and limited preemption for run-to-completion parallel processor scheduling policies

    Performance Evaluation Rev.

    (1994)
  • D. Eager et al.

    Speedup versus efficiency in parallel systems

    IEEE Trans. Comput.

    (March 1989)
  • H. El-Rewini et al.

    Task Scheduling in Parallel and Distributed Systems

    (1994)
  • D. Ghosal et al.

    The processor working set and its use in scheduling multiprocessor systems

    IEEE Trans. Software Eng.

    (May 1991)
  • A. Gupta, Analysis and design of scalable parallel algorithms for scientific computing, Ph.D. Thesis, University of...
  • K. Hwang

    Advanced Computer Architecture: Parallelism, Scalability, Programmability

    (1993)
  • K. Jeffay et al.

    On non-preemptive scheduling of periodic and sporadic tasks

    Proc. IEEE Real-Time Systems Symp.

    (1994)
  • Cited by (26)

    • Requirement-aware strategies for scheduling real-time divisible loads on clusters

      2013, Journal of Parallel and Distributed Computing
      Citation Excerpt :

      The paper [17] presented a dynamic multiprocessor real-time scheduling algorithm which allocates a task to multiple processors when the allocation of one processor cannot guarantee the deadline. The authors in [11] proposed a dynamic scalable task scheduling algorithm, in which the number of processors allocated to each task is kept as small as possible while the deadline is satisfied and a task with a large workload derivative is favored to start earlier. The aforementioned works assumed that when a task is parallelized, all its parallel subtasks have to start at the same time in order to synchronize their executions.

    • Efficient real-time divisible load scheduling

      2012, Journal of Parallel and Distributed Computing
      Citation Excerpt :

      However, without a good QoS mechanism, current cluster sites cannot provide these users with good response time estimations. Real-time divisible load scheduling is a well-researched area [8,9,17–20], where researchers investigate how to provide time predictability to the aforementioned divisible load applications. Focusing on satisfying QoS, providing time guarantees, and better utilizing cluster resources, existing approaches give little emphasis to scheduling efficiency.

    • Real-time scheduling of divisible loads in cluster computing environments

      2010, Journal of Parallel and Distributed Computing
      Citation Excerpt :

      Because bag-of-tasks applications are not arbitrarily divisible, they are different from the divisible loads investigated in this paper. The most closely related work [23] to ours is scheduling algorithms for “scalable real-time tasks” running on a multiprocessor system. In that paper, like divisible loads, it is assumed that a task can be executed on more than one processor, and as more processors are allocated, its pure computation time decreases monotonically.

    • Unavailable time aware scheduling of hybrid task on heterogeneous distributed system

      2020, IAENG International Journal of Applied Mathematics
    View all citing articles on Scopus

    This work was supported by the Research Grant from Hallym university, Korea and the Ministry of Education of Korea through its BK21 program toward the Electrical and Computer Engineering Division at POSTECH.

    View full text