skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Characterizing the Impact of Soft Errors Affecting Floating-point ALUs using RTL-level Fault Injection

Abstract

Strategies to detect, correct, or mitigate the impact of soft errors rely on errors-injection experiments. For efficient evaluation, such experiments typically inject errors in software by sampling errors from a candidate distribution. Most often, these strategies randomly select and flip one bit in the output of an instruction. While single-bit flips might constitute a meaningful model for errors affecting hardware, the appropriateness of this model for software-based errors has not been studied. In this paper, we study the manifestation of errors in the output registers due to errors affecting candidate instructions executed by floating point ALUs. We inject single-bit flips into the RTL descriptions of floating point ALUs and analyze the differences between anticipated and observed outputs when executing floating-point addition, subtraction, multiplication, and division. We choose the operands for these instructions randomly and from operands observed in five benchmarks. We observe a rich distribution of errors in the output and analyze their implications for software-based fault-injection campaigns.

Authors:
ORCiD logo [1];  [2];  [2];  [1]
  1. BATTELLE (PACIFIC NW LAB)
  2. University of Texas at Austin
Publication Date:
Research Org.:
Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
Sponsoring Org.:
USDOE
OSTI Identifier:
1617875
Report Number(s):
PNNL-SA-134868
DOE Contract Number:  
AC05-76RL01830
Resource Type:
Conference
Resource Relation:
Conference: Proceedings of the 47th International Conference on Parallel Processing, (ICPP 2018), August 13-16, 2018, Eugene, OR
Country of Publication:
United States
Language:
English

Citation Formats

Subasi, Omer, Chang, Chunkai, Erez, Mattan, and Krishnamoorthy, Sriram. Characterizing the Impact of Soft Errors Affecting Floating-point ALUs using RTL-level Fault Injection. United States: N. p., 2018. Web.
Subasi, Omer, Chang, Chunkai, Erez, Mattan, & Krishnamoorthy, Sriram. Characterizing the Impact of Soft Errors Affecting Floating-point ALUs using RTL-level Fault Injection. United States.
Subasi, Omer, Chang, Chunkai, Erez, Mattan, and Krishnamoorthy, Sriram. Mon . "Characterizing the Impact of Soft Errors Affecting Floating-point ALUs using RTL-level Fault Injection". United States.
@article{osti_1617875,
title = {Characterizing the Impact of Soft Errors Affecting Floating-point ALUs using RTL-level Fault Injection},
author = {Subasi, Omer and Chang, Chunkai and Erez, Mattan and Krishnamoorthy, Sriram},
abstractNote = {Strategies to detect, correct, or mitigate the impact of soft errors rely on errors-injection experiments. For efficient evaluation, such experiments typically inject errors in software by sampling errors from a candidate distribution. Most often, these strategies randomly select and flip one bit in the output of an instruction. While single-bit flips might constitute a meaningful model for errors affecting hardware, the appropriateness of this model for software-based errors has not been studied. In this paper, we study the manifestation of errors in the output registers due to errors affecting candidate instructions executed by floating point ALUs. We inject single-bit flips into the RTL descriptions of floating point ALUs and analyze the differences between anticipated and observed outputs when executing floating-point addition, subtraction, multiplication, and division. We choose the operands for these instructions randomly and from operands observed in five benchmarks. We observe a rich distribution of errors in the output and analyze their implications for software-based fault-injection campaigns.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {2018},
month = {8}
}

Conference:
Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share: