SaFirE: Scalable and accurate fault injection for parallel multithreaded applications

Giorgis Georgakoudis; Ignacio Laguna; Hans Vandierendonck; Dimitrios S. Nikolopoulos; Martin Schulz

doi:10.1109/IPDPS.2019.00097

SaFirE: Scalable and accurate fault injection for parallel multithreaded applications

Giorgis Georgakoudis, Ignacio Laguna, Hans Vandierendonck, Dimitrios S. Nikolopoulos, Martin Schulz

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

1 Citation (Scopus)

Abstract

Soft errors threaten to disrupt supercomputing scaling. Fault injection is a key technique to understand the impact of faults on scientific applications. However, injecting faults in parallel applications has been prohibitively slow, inaccurate and hard to implement. In this paper, we present SAFIRE, the first fast and accurate fault injection framework for parallel, multi-threaded applications. SAFIRE uses novel compiler instrumentation and code generation techniques to achieve high accuracy and high speed. Using SAFIRE, we show that fault manifestations can be significantly different depending on whether they happen in the application itself or in the parallel runtime system. In our experimental evaluation on 15 HPC parallel programs, we show that SAFIRE is multiple factors faster and equally accurate in comparison with state-of-the-art dynamic binary instrumentation tools for fault injection.

Original language	English
Title of host publication	Proceedings - 2019 IEEE 33rd International Parallel and Distributed Processing Symposium, IPDPS 2019
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	890-899
Number of pages	10
ISBN (Electronic)	9781728112466
DOIs	https://doi.org/10.1109/IPDPS.2019.00097
Publication status	Published - 01 May 2019
Event	33rd IEEE International Parallel and Distributed Processing Symposium, IPDPS 2019 - Rio de Janeiro, Brazil Duration: 20 May 2019 → 24 May 2019

Conference

Conference	33rd IEEE International Parallel and Distributed Processing Symposium, IPDPS 2019
Country	Brazil
City	Rio de Janeiro
Period	20/05/2019 → 24/05/2019

ASJC Scopus subject areas

Computer Networks and Communications
Hardware and Architecture
Information Systems and Management

Access to Document

10.1109/IPDPS.2019.00097

Link to publication in Scopus

Fingerprint Dive into the research topics of 'SaFirE: Scalable and accurate fault injection for parallel multithreaded applications'. Together they form a unique fingerprint.

Projects

1 Finished
1 Active

R6551CSC: Open TransPREcision COMPuting

Woods, R., Karakonstantis, G. & Vandierendonck, H.

03/11/2016 → …

Project: Research

R1485CSC: SERT: Scale-free, Energy-Aware and Resilient Adaptation of CSE Applications to Mega-Core Systems

Nikolopoulos, D., Scott, S., Vandierendonck, H. & de Supinski, B.

13/11/2014 → 30/09/2018

Project: Research

Cite this

Georgakoudis, G., Laguna, I., Vandierendonck, H., Nikolopoulos, D. S., & Schulz, M. (2019). SaFirE: Scalable and accurate fault injection for parallel multithreaded applications. In Proceedings - 2019 IEEE 33rd International Parallel and Distributed Processing Symposium, IPDPS 2019 (pp. 890-899). [8820954] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IPDPS.2019.00097

Georgakoudis, Giorgis ; Laguna, Ignacio ; Vandierendonck, Hans ; Nikolopoulos, Dimitrios S. ; Schulz, Martin. / SaFirE : Scalable and accurate fault injection for parallel multithreaded applications. Proceedings - 2019 IEEE 33rd International Parallel and Distributed Processing Symposium, IPDPS 2019. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 890-899

@inproceedings{82c61b80b9934b23836ac11fbbc6b366,

title = "SaFirE: Scalable and accurate fault injection for parallel multithreaded applications",

abstract = "Soft errors threaten to disrupt supercomputing scaling. Fault injection is a key technique to understand the impact of faults on scientific applications. However, injecting faults in parallel applications has been prohibitively slow, inaccurate and hard to implement. In this paper, we present SAFIRE, the first fast and accurate fault injection framework for parallel, multi-threaded applications. SAFIRE uses novel compiler instrumentation and code generation techniques to achieve high accuracy and high speed. Using SAFIRE, we show that fault manifestations can be significantly different depending on whether they happen in the application itself or in the parallel runtime system. In our experimental evaluation on 15 HPC parallel programs, we show that SAFIRE is multiple factors faster and equally accurate in comparison with state-of-the-art dynamic binary instrumentation tools for fault injection.",

author = "Giorgis Georgakoudis and Ignacio Laguna and Hans Vandierendonck and Nikolopoulos, {Dimitrios S.} and Martin Schulz",

year = "2019",

month = may,

day = "1",

doi = "10.1109/IPDPS.2019.00097",

language = "English",

pages = "890--899",

booktitle = "Proceedings - 2019 IEEE 33rd International Parallel and Distributed Processing Symposium, IPDPS 2019",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

note = "33rd IEEE International Parallel and Distributed Processing Symposium, IPDPS 2019 ; Conference date: 20-05-2019 Through 24-05-2019",

}

Georgakoudis, G, Laguna, I, Vandierendonck, H , Nikolopoulos, DS & Schulz, M 2019, SaFirE: Scalable and accurate fault injection for parallel multithreaded applications. in Proceedings - 2019 IEEE 33rd International Parallel and Distributed Processing Symposium, IPDPS 2019., 8820954, Institute of Electrical and Electronics Engineers Inc., pp. 890-899, 33rd IEEE International Parallel and Distributed Processing Symposium, IPDPS 2019, Rio de Janeiro, Brazil, 20/05/2019. https://doi.org/10.1109/IPDPS.2019.00097

SaFirE : Scalable and accurate fault injection for parallel multithreaded applications. / Georgakoudis, Giorgis; Laguna, Ignacio; Vandierendonck, Hans ; Nikolopoulos, Dimitrios S.; Schulz, Martin.

Proceedings - 2019 IEEE 33rd International Parallel and Distributed Processing Symposium, IPDPS 2019. Institute of Electrical and Electronics Engineers Inc., 2019. p. 890-899 8820954.

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

TY - GEN

T1 - SaFirE

T2 - 33rd IEEE International Parallel and Distributed Processing Symposium, IPDPS 2019

AU - Georgakoudis, Giorgis

AU - Laguna, Ignacio

AU - Vandierendonck, Hans

AU - Nikolopoulos, Dimitrios S.

AU - Schulz, Martin

PY - 2019/5/1

Y1 - 2019/5/1

N2 - Soft errors threaten to disrupt supercomputing scaling. Fault injection is a key technique to understand the impact of faults on scientific applications. However, injecting faults in parallel applications has been prohibitively slow, inaccurate and hard to implement. In this paper, we present SAFIRE, the first fast and accurate fault injection framework for parallel, multi-threaded applications. SAFIRE uses novel compiler instrumentation and code generation techniques to achieve high accuracy and high speed. Using SAFIRE, we show that fault manifestations can be significantly different depending on whether they happen in the application itself or in the parallel runtime system. In our experimental evaluation on 15 HPC parallel programs, we show that SAFIRE is multiple factors faster and equally accurate in comparison with state-of-the-art dynamic binary instrumentation tools for fault injection.

AB - Soft errors threaten to disrupt supercomputing scaling. Fault injection is a key technique to understand the impact of faults on scientific applications. However, injecting faults in parallel applications has been prohibitively slow, inaccurate and hard to implement. In this paper, we present SAFIRE, the first fast and accurate fault injection framework for parallel, multi-threaded applications. SAFIRE uses novel compiler instrumentation and code generation techniques to achieve high accuracy and high speed. Using SAFIRE, we show that fault manifestations can be significantly different depending on whether they happen in the application itself or in the parallel runtime system. In our experimental evaluation on 15 HPC parallel programs, we show that SAFIRE is multiple factors faster and equally accurate in comparison with state-of-the-art dynamic binary instrumentation tools for fault injection.

UR - http://www.scopus.com/inward/record.url?scp=85072822375&partnerID=8YFLogxK

U2 - 10.1109/IPDPS.2019.00097

DO - 10.1109/IPDPS.2019.00097

M3 - Conference contribution

AN - SCOPUS:85072822375

SP - 890

EP - 899

BT - Proceedings - 2019 IEEE 33rd International Parallel and Distributed Processing Symposium, IPDPS 2019

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 20 May 2019 through 24 May 2019

ER -

Georgakoudis G, Laguna I, Vandierendonck H , Nikolopoulos DS, Schulz M. SaFirE: Scalable and accurate fault injection for parallel multithreaded applications. In Proceedings - 2019 IEEE 33rd International Parallel and Distributed Processing Symposium, IPDPS 2019. Institute of Electrical and Electronics Engineers Inc. 2019. p. 890-899. 8820954 https://doi.org/10.1109/IPDPS.2019.00097