Understanding the Effects of Data Corruption on Application Behavior Based on Data Characteristics

George  Stefanakis; Vijay Nagarajan; Marcelo Cintra

doi:10.1007/978-3-319-24255-2_12

Understanding the Effects of Data Corruption on Application Behavior Based on Data Characteristics

George Stefanakis, Vijay Nagarajan, Marcelo Cintra

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

In this paper, the results of an experimental study on the error sensitivities of application data are presented. We develop a portable software-implemented fault-injection (SWIFI) tool that, on top of performing single-bit
flip fault injections and capturing their eects on application behavior, is also data-level aware and tracks the corrupted application data to report their high-level characteristics (usage type, size, user, memory space location). After extensive testing of NPB-serial (7.8M fault injections), we are able to characterize the sensitivities of data based on their high-level characteristics. Moreover, we conclude that application data are error sensitive in parts; depending on their type, they have distinct and wide less-sensitive bit ranges either at the MSBs or LSBs. Among other uses, such gained insight could drive the development of sensitivity-aware protection mechanisms of application
data .

Original language	English
Title of host publication	Computer Safety, Reliability, and Security
Subtitle of host publication	34th International Conference, SAFECOMP 2015, Delft, The Netherlands, September 23-25, 2015, Proceedings
Publisher	Springer International Publishing
Pages	151-165
Number of pages	15
ISBN (Electronic)	978-3-319-24255-2
ISBN (Print)	978-3-319-24254-5
DOIs	https://doi.org/10.1007/978-3-319-24255-2_12
Publication status	Published - 26 Sep 2015

Publication series

Name	Lecture Notes in Computer Science
Volume	9337
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Access to Document

10.1007/978-3-319-24255-2_12

safecomp15Accepted author manuscript, 1.47 MB

http://link.springer.com/chapter/10.1007/978-3-319-24255-2_12

1 Finished

Error-tolerant Stream Processing System Design (ESP-SD)
Nagarajan, V. & Grot, B.
EPSRC
30/11/14 → 29/11/17
Project: Research

Cite this

Stefanakis, G., Nagarajan, V., & Cintra, M. (2015). Understanding the Effects of Data Corruption on Application Behavior Based on Data Characteristics. In Computer Safety, Reliability, and Security: 34th International Conference, SAFECOMP 2015, Delft, The Netherlands, September 23-25, 2015, Proceedings (pp. 151-165). (Lecture Notes in Computer Science; Vol. 9337). Springer International Publishing. https://doi.org/10.1007/978-3-319-24255-2_12

Stefanakis, George ; Nagarajan, Vijay ; Cintra, Marcelo. / Understanding the Effects of Data Corruption on Application Behavior Based on Data Characteristics. Computer Safety, Reliability, and Security: 34th International Conference, SAFECOMP 2015, Delft, The Netherlands, September 23-25, 2015, Proceedings. Springer International Publishing, 2015. pp. 151-165 (Lecture Notes in Computer Science).

@inproceedings{a603b6a9d7bb42e8b3437e307074618e,

title = "Understanding the Effects of Data Corruption on Application Behavior Based on Data Characteristics",

abstract = "In this paper, the results of an experimental study on the error sensitivities of application data are presented. We develop a portable software-implemented fault-injection (SWIFI) tool that, on top of performing single-bit flip fault injections and capturing their eects on application behavior, is also data-level aware and tracks the corrupted application data to report their high-level characteristics (usage type, size, user, memory space location). After extensive testing of NPB-serial (7.8M fault injections), we are able to characterize the sensitivities of data based on their high-level characteristics. Moreover, we conclude that application data are error sensitive in parts; depending on their type, they have distinct and wide less-sensitive bit ranges either at the MSBs or LSBs. Among other uses, such gained insight could drive the development of sensitivity-aware protection mechanisms of applicationdata .",

author = "George Stefanakis and Vijay Nagarajan and Marcelo Cintra",

year = "2015",

month = sep,

day = "26",

doi = "10.1007/978-3-319-24255-2_12",

language = "English",

isbn = "978-3-319-24254-5",

series = "Lecture Notes in Computer Science",

publisher = "Springer International Publishing",

pages = "151--165",

booktitle = "Computer Safety, Reliability, and Security",

}

Stefanakis, G, Nagarajan, V & Cintra, M 2015, Understanding the Effects of Data Corruption on Application Behavior Based on Data Characteristics. in Computer Safety, Reliability, and Security: 34th International Conference, SAFECOMP 2015, Delft, The Netherlands, September 23-25, 2015, Proceedings. Lecture Notes in Computer Science, vol. 9337, Springer International Publishing, pp. 151-165. https://doi.org/10.1007/978-3-319-24255-2_12

Understanding the Effects of Data Corruption on Application Behavior Based on Data Characteristics. / Stefanakis, George ; Nagarajan, Vijay; Cintra, Marcelo.

Computer Safety, Reliability, and Security: 34th International Conference, SAFECOMP 2015, Delft, The Netherlands, September 23-25, 2015, Proceedings. Springer International Publishing, 2015. p. 151-165 (Lecture Notes in Computer Science; Vol. 9337).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Understanding the Effects of Data Corruption on Application Behavior Based on Data Characteristics

AU - Stefanakis, George

AU - Nagarajan, Vijay

AU - Cintra, Marcelo

PY - 2015/9/26

Y1 - 2015/9/26

N2 - In this paper, the results of an experimental study on the error sensitivities of application data are presented. We develop a portable software-implemented fault-injection (SWIFI) tool that, on top of performing single-bit flip fault injections and capturing their eects on application behavior, is also data-level aware and tracks the corrupted application data to report their high-level characteristics (usage type, size, user, memory space location). After extensive testing of NPB-serial (7.8M fault injections), we are able to characterize the sensitivities of data based on their high-level characteristics. Moreover, we conclude that application data are error sensitive in parts; depending on their type, they have distinct and wide less-sensitive bit ranges either at the MSBs or LSBs. Among other uses, such gained insight could drive the development of sensitivity-aware protection mechanisms of applicationdata .

AB - In this paper, the results of an experimental study on the error sensitivities of application data are presented. We develop a portable software-implemented fault-injection (SWIFI) tool that, on top of performing single-bit flip fault injections and capturing their eects on application behavior, is also data-level aware and tracks the corrupted application data to report their high-level characteristics (usage type, size, user, memory space location). After extensive testing of NPB-serial (7.8M fault injections), we are able to characterize the sensitivities of data based on their high-level characteristics. Moreover, we conclude that application data are error sensitive in parts; depending on their type, they have distinct and wide less-sensitive bit ranges either at the MSBs or LSBs. Among other uses, such gained insight could drive the development of sensitivity-aware protection mechanisms of applicationdata .

U2 - 10.1007/978-3-319-24255-2_12

DO - 10.1007/978-3-319-24255-2_12

M3 - Conference contribution

SN - 978-3-319-24254-5

T3 - Lecture Notes in Computer Science

SP - 151

EP - 165

BT - Computer Safety, Reliability, and Security

PB - Springer International Publishing

ER -

Stefanakis G, Nagarajan V, Cintra M. Understanding the Effects of Data Corruption on Application Behavior Based on Data Characteristics. In Computer Safety, Reliability, and Security: 34th International Conference, SAFECOMP 2015, Delft, The Netherlands, September 23-25, 2015, Proceedings. Springer International Publishing. 2015. p. 151-165. (Lecture Notes in Computer Science). https://doi.org/10.1007/978-3-319-24255-2_12

Understanding the Effects of Data Corruption on Application Behavior Based on Data Characteristics

Abstract

Publication series

Access to Document

Fingerprint

Projects

Error-tolerant Stream Processing System Design (ESP-SD)

Cite this