Feasibility of Fork-Join Real-Time Task Graph Models

Author:

Sun Jinghao1,Guan Nan1,Wang Yang1,Deng Qingxu1,Zeng Peng2,Yi Wang3

Affiliation:

1. Northeastern University, China

2. Chinese Academy of Sciences, China

3. Uppsala University, Sweden

Abstract

In the formal analysis of real-time systems, modeling of branching codes and modeling of intratask parallelism structures are two of the most important research topics. These two real-time properties are combined, resulting in the fork-join real-time task (FJRT) model, which extends the digraph-based task model with forking and joining semantics. We prove that the EDF schedulability problem on a preemptive uniprocessor for the FJRT model is coNP-hard in the strong sense, even if the utilization of the task system is bounded by a constant strictly less than 1. Then, we show that the problem becomes tractable with some slight structural restrictions on parallel sections, for which we propose an exact schedulability test with pseudo-polynomial time complexity. Our results thus establish a borderline between the tractable and intractable FJRT models.

Funder

Natural Science Foundation of China

National Basic Pre-research Program of China

Publisher

Association for Computing Machinery (ACM)

Subject

Hardware and Architecture,Software

Cited by 12 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. Real-Time Task Models;Handbook of Real-Time Computing;2022

2. Simultaneous Scheduling and Core-Type Optimization for Moldable Fork-Join Tasks on Heterogeneous Multicores;IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences;2021

3. Algorithms for Computing the WCRT Bound of OpenMP Task Systems With Conditional Branches;IEEE Transactions on Computers;2021-01-01

4. Efficient Feasibility Analysis for Graph-Based Real-Time Task Systems;IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems;2020-11

5. Scheduling of moldable fork-join tasks with inter- and intra-task communications;Proceedings of the 23th International Workshop on Software and Compilers for Embedded Systems;2020-05-25

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3