Loading...

Analysis of Software-based Techniques for Detecting Soft Errors in Embedded Systems

Ghasemi, Golsa | 2014

704 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 46221 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Miremadi, Ghassem
  7. Abstract:
  8. The development of embedded systems in safety critical applications leads to using fault tolerant methods to improve their dependability. Among the various fault tolerant methods, recently, software based methods are more appealing during to the following reasons: 1) They are applicable to the Commercial-Off-The Shelf (COTS) processors and lead to reduction in cost and time to market, 2) Unlike the traditional hardware based internal error detection methods, software based methods fault coverage are more promising, and 3)Software-based methods are cheaper, quicker and do not require any hardware modifications. The technology scaling trend in recent years, made system more vulnerable to various fault models; i.e. radiation induced soft errors. Regardless of the methods used to improve the reliability of system, this trend force traditional methods to impose significant penalties to the system; since they uniformly protect different program regions. Since each part of the program has various vulnerabilities to faults, in recent years a vulnerability factor which was called IVF (instruction vulnerability factor) introduced to alleviate the performance overhead of traditional software based fault tolerant methods. In this thesis, three software fault tolerance methods are implemented. Then, a software-based technique to detect and mask the program faults is proposed. All four mentioned methods are simulated on LEON2 embedded processor and evaluated by fault injection. The experimental results on MiBench benchmarks suite show that the performance and code size overhead of software-based techniques can be reduced by using vulnerability factors. It should be note that, the proposed method incurs 26% performance overhead while it can mask 89% of faults in protected instructions of the applications
  9. Keywords:
  10. Safety-Critical Application ; Embedded System ; Software Fault Tolerance ; Software-based Techniques For Reliability Improvement

 Digital Object List

 Bookmark

No TOC