• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文OA文献 >Efficient fault-injection-based assessment of software-implemented hardware fault tolerance
【2h】

Efficient fault-injection-based assessment of software-implemented hardware fault tolerance

机译：基于故障注入的有效软件评估硬件容错能力评估

页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

With continuously shrinking semiconductor structure sizes and lower supplyvoltages, the per-device susceptibility to transient and permanent hardwarefaults is on the rise. A class of countermeasures with growing popularityis Software-Implemented Hardware Fault Tolerance (SIHFT), which avoidsexpensive hardware mechanisms and can be applied application-specifically.However, SIHFT can, against intuition, cause more harm than good, becauseits overhead in execution time and memory space also increases the figurative“attack surface” of the system – it turns out that application-specific configuration of SIHFT is in fact a necessity rather than just an advantage.Consequently, target programs need to be analyzed for particularly critical spots to harden. SIHFT-hardened programs need to be measured and compared throughout all development phases of the program to observe reliability improvements or deteriorations over time. Additionally, SIHFT implementationsneed to be tested.The contributions of this dissertation focus on Fault Injection (FI) as an assessment technique satisfying all these requirements – analysis, measurement and comparison, and test. I describe the design and implementation of an FI tool, named Fail*, that overcomes several shortcomings in the state ofthe art, and enables research on the general drawbacks of simulation-basedFI. As demonstrated in four case studies in the context of SIHFT research,Fail* provides novel fine-grained analysis techniques that exploit the newlygained possibility to analyze FI results from complete fault-space exploration.These analysis techniques aid SIHFT design decisions on the level of programmodules, functions, variables, source-code lines, or single machine instructions.Based on the experience from the case studies, I address the problemof large computation efforts that accompany exhaustive fault-space explorationfrom two different angles: Firstly, I develop a heuristical fault-spacepruning technique that allows to freely trade the total FI-experiment countfor result accuracy, while still providing information on all possible faultspacecoordinates. Secondly, I speed up individual TAP-based FI experimentsby improving the fast-forwarding operation by several orders of magnitudefor most workloads. Finally, I dissect current practices in FI-based evaluationof SIHFT-hardened programs, identify three widespread pitfalls in theresult interpretation, and advance the state of the art by defining a novelcomparison metric.
机译：随着半导体结构尺寸的不断缩小和较低的电源电压，每个器件对瞬态和永久性硬件故障的敏感性正在上升。一类日益流行的对策是软件实现的硬件容错（SIHFT），它避免了昂贵的硬件机制，可以在特定的应用程序中应用。但是，由于直觉，SIHFT带来的弊大于利，因为它在执行时间和内存上的开销空间还增加了系统的象征性“攻击面” –事实证明，针对应用程序进行特定配置的SIHFT实际上是必要的，而不仅仅是优点。因此，需要分析目标程序以发现特别关键的地方以进行硬化。需要对SIHFT强化程序进行度量，并在该程序的所有开发阶段进行比较，以观察可靠性随时间的提高或降低。此外，还需要对SIHFT实现进行测试。本论文的重点是故障注入（FI）作为一种评估技术，可以满足所有这些要求–分析，测量，比较和测试。我描述了一种名为Fail *的FI工具的设计和实现，该工具克服了现有技术中的几个缺点，并使得能够研究基于仿真的FI的一般缺点。正如在SIHFT研究背景下进行的四个案例研究中所证明的那样，Fail *提供了新颖的细粒度分析技术，该技术利用新获得的可能性来分析来自完整故障空间探索的FI结果。这些分析技术有助于在程序模块级别上进行SIHFT设计决策。 ，函数，变量，源代码行或单个机器指令。基于案例研究的经验，我从两个不同的角度解决了伴随详尽的故障空间探索而进行的大量计算工作：首先，我开发了启发式故障-空间修剪技术，可以自由地交换FI实验总数以获得结果准确性，同时仍然提供所有可能的故障空间坐标信息。其次，我通过将大多数工作负载的快进操作提高了几个数量级，从而加快了基于TAP的FI实验的速度。最后，我剖析了基于FI的SIHFT强化程序评估中的当前实践，确定了结果解释中的三个普遍存在的陷阱，并通过定义一种新颖的比较指标来提高了技术水平。

著录项

  • 作者

    Schirmeier Horst Benjamin;

  • 作者单位
  • 年度 2016
  • 总页数
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类

相似文献

  • 外文文献
  • 中文文献
  • 专利

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