首页> 外文期刊>Journal of supercomputing >Transformation process of RTS scheduling analysis requirements from UML/MARTE to dynamic priority time Petri Nets
【24h】

Transformation process of RTS scheduling analysis requirements from UML/MARTE to dynamic priority time Petri Nets

机译:RTS调度分析要求从UML / MARTE到动态优先级时间Petri网的转换过程

获取原文
获取原文并翻译 | 示例

摘要

The real-time system (RTS) design is a major challenge due to the complexity of the considered application and the appearance of multiprocessor architectures. Most of the research in this area is interested in high abstraction level methods to decrease the design convolution. In particular, the Unified Modeling Language (UML) profiles and the Model Driven Engineering (MDE) aim at being an adequate solution to support the whole life cycle of RTS design with its real-time constraints and performance issues. Based on MDE and the Modeling and Analysis of Real-Time and Embedded systems (MARTE) profile, the present manuscript proposes an interactive partitioning scheduling of RTS running on multiprocessor architecture. A special attention is paid to the scheduling analysis step to accelerate the exploration of HW/SW space solution. Starting from a candidate solution modeled with UML activity and annotated with MARTE stereotypes, a mapping process to dynamic priority time Petri Nets (dPTPN) was defined. The considered dPTPN is able to prove the schedulability or provide a counterexample to determine the partitions causes of temporal fault. This counterexample presents a useful feedback to the HW/SW space solution explorer in the aim to exclude all solution containing the described partitions.
机译:由于所考虑的应用程序的复杂性和多处理器体系结构的出现,实时系统(RTS)设计是一项重大挑战。该领域的大多数研究都对降低设计卷积的高抽象级别方法感兴趣。特别是,统一建模语言(UML)概要文件和模型驱动工程(MDE)旨在通过其实时约束和性能问题,为支持RTS设计的整个生命周期提供适当的解决方案。本文基于MDE以及实时和嵌入式系统的建模与分析(MARTE)配置文件,提出了在多处理器体系结构上运行的RTS的交互式分区调度。特别注意调度分析步骤,以加快对硬件/软件空间解决方案的探索。从以UML活动为模型并以MARTE构造型进行注释的候选解决方案开始,定义了到动态优先级时间Petri网(dPTPN)的映射过程。所考虑的dPTPN能够证明可调度性或提供反例以确定临时故障的分区原因。该反例向HW / SW空间解决方案浏览器提供了有用的反馈,旨在排除所有包含所描述分区的解决方案。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号