ISSN 1000-1239 CN 11-1777/TP

• 软件技术 •

### 通信竞争的混合关键级系统多DAG动态调度策略

1. 1(湖南大学信息科学与工程学院 长沙 410082); 2(湖南大学嵌入式与网络计算湖南省重点实验室 长沙 410082); 3(湖南省发展和改革委员会 长沙 410004); 4(厦门理工学院计算机与信息工程学院 福建厦门 361024) (llj1984109@qq.com)
• 出版日期: 2015-11-01
• 基金资助:
基金项目：国家自然科学基金项目(61173036,61202102,61300039,61300037,61502405);国家“八六三”高技术研究发展计划基金项目(2012AA01A301-01)

### Multiple DAGs Dynamic Scheduling for Mixed-Criticality Systems with Communication Contention

Liu Liangjiao1,2, Xie Guoqi1,2, Li Renfa1,2, Yang Liu3, Xie Yong4

1. 1(College of Computer Science and Electronic Engineering, Hunan University, Changsha 410082);2(Key Laboratory for Embedded and Network Computing of Hunan Province, Hunan University, Changsha 410082);3(Development and Reform Commission of Hunan Province, Changsha 410004);4(College of Computer and Information Engineering, Xiamen University of Technology, Xiamen, Fujian 361024)
• Online: 2015-11-01

Abstract: The scheduling of mixed-criticality systems with communication contention based on multiple DAGs model is the requirement of automobile electronic systems with heterogeneity and distribution. Firstly, the more accurate “upward rank value” and “earliest finish time” with communication contention on time are implemented to adapt to the heterogeneity of both computing and networking, and the synchronization of task and message in this paper. Then, an algorithm called fairness on multiple DAGs dynamic task and message scheduling (F_MDDTMS) is proposed to reduce scheduling length. An algorithm called criticality on multiple DAGs dynamic task and message scheduling (C_MDDTMS) is proposed to ensure the real-time of higher criticality applications. An algorithm called mixed criticality on multiple DAGs dynamic task and message scheduling (MC_MDDTMS) is also proposed based on the joint of F_MDDTMS algorithm and C_MDDTMS algorithm to ensure hard real-time of higher criticality applications and the activity of lower criticality applications of mixed-criticality systems. Example analysis and experimental results show that the proposed algorithms are excellent on scheduling length, unfairness, worst-case response time (WCRT) and real-time.