ITR: Collaborative Research: SPARTA: Static Parametric Timing Analysis to Support Dynamic Decisions inEmbedded Systems

ITR：协作研究：SPARTA：支持嵌入式系统动态决策的静态参数时序分析

• 批准号: 0312493
• 负责人: David Whalley
• 金额: $ 13万
• 依托单位: Florida State University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2003
• 资助国家: 美国
• 起止时间: 2003-09-01 至 2007-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0312493&HistoricalAwards=false
• 关键词: ITR; Collaborative; Research; SPARTA; Static

MuellerWhalleyHealy Embedded systems with temporal constraints rely on timely schedulingand a priori knowledge of worst-case execution times. Static timinganalysis derives safe bounds of WCETs but its applicability has beenlimited to hard real-time systems and small code snippets.This research addresses these limitations of timing analysis forembedded systems. It contributes a novel approach to program analysisthrough parametric techniques of static timing analysis and providesinnovative methods for exploiting them. The proposed techniquesexpress worst-case execution time as a formula with the followingbenefits. First, static timing analysis becomes applicable to a wideclass of embedded systems with variable loop bounds. Second, thebenefits of parametric timing analysis provide new opportunities fordynamic and flexible scheduling decisions. Third, timing abstractionsthrough the parametric approach enable the analysis of much largerprograms than in the past. Finally, the techniques of static timinganalysis, currently constrained to hard real-time system, becomeapplicable to a much wider range of embedded systems with soft timingconstraints. The broader impact is to increasingly expose students toembedded systems and to provide essential temporal assurances, whichare a prerequisite for applying the results of hard and soft real-timescheduling for reliable embedded systems of national and economic importance.

MuellerWhalleyHealy 具有时间约束的嵌入式系统依赖于及时调度和最坏情况执行时间的先验知识。 静态时序分析得出 WCET 的安全范围，但其适用性仅限于硬实时系统和小代码片段。本研究解决了嵌入式系统时序分析的这些局限性。它通过静态时序分析的参数技术提供了一种新颖的程序分析方法，并提供了利用它们的创新方法。 所提出的技术将最坏情况的执行时间表示为具有以下优点的公式。首先，静态时序分析适用于具有可变循环边界的各种嵌入式系统。其次，参数化时序分析的好处为动态、灵活的调度决策提供了新的机会。第三，通过参数化方法进行时序抽象可以分析比过去大得多的程序。 最后，目前仅限于硬实时系统的静态时序分析技术可适用于更广泛的具有软时序约束的嵌入式系统。更广泛的影响是让学生越来​​越多地接触嵌入式系统并提供必要的时间保证，这是将硬实时调度和软实时调度的结果应用于具有国家和经济重要性的可靠嵌入式系统的先决条件。