Evaluating user-level fault tolerance for MPI applications

Ignacio Laguna; David F. Richards; Todd Gamblin; Martin Schulz; Bronis R. De Supinski

doi:10.1145/2642769.2642775

Evaluating user-level fault tolerance for MPI applications

Ignacio Laguna, David F. Richards, Todd Gamblin, Martin Schulz, Bronis R. De Supinski

Lawrence Livermore National Laboratory

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

28 Scopus citations

Abstract

The User Level Failure Mitigation (ULFM) interface has been proposed to provide fault-tolerant semantics in MPI. Previous work has presented performance evaluations of the interface; yet questions related to its programability and applicability remain unanswered. In this paper, we present our experiences on using ULFM in a case study (a large molecular dynamics application) to shed light on the advantages and difficulties of this interface to program fault-tolerant MPI applications. We found that, although ULFM is suitable for applications with work-decomposition flexibility (e.g., master-slave), it provides few benefits for more general (e.g., bulk synchronous) MPI applications.

Original language	English
Title of host publication	Proceedings of the 21st European MPI Users' Group Meeting, EuroMPI/ASIA 2014
Publisher	Association for Computing Machinery
Pages	57-62
Number of pages	6
ISBN (Electronic)	9781450328753
DOIs	https://doi.org/10.1145/2642769.2642775
State	Published - 9 Sep 2014
Externally published	Yes
Event	21st European MPI Users' Group Meeting, EuroMPI/ASIA 2014 - Kyoto, Japan Duration: 9 Sep 2014 → 12 Sep 2014

Publication series

Name	ACM International Conference Proceeding Series
Volume	09-12-September-2014

Conference

Conference	21st European MPI Users' Group Meeting, EuroMPI/ASIA 2014
Country/Territory	Japan
City	Kyoto
Period	9/09/14 → 12/09/14

Keywords

Failure receovery models
Fault tolerance
MPI
Molecular dynamics simulation

Access to Document

10.1145/2642769.2642775

Cite this

Laguna, I., Richards, D. F., Gamblin, T., Schulz, M., & De Supinski, B. R. (2014). Evaluating user-level fault tolerance for MPI applications. In Proceedings of the 21st European MPI Users' Group Meeting, EuroMPI/ASIA 2014 (pp. 57-62). (ACM International Conference Proceeding Series; Vol. 09-12-September-2014). Association for Computing Machinery. https://doi.org/10.1145/2642769.2642775

@inproceedings{87b72fc485cb429697f2040a2dcf1774,

title = "Evaluating user-level fault tolerance for MPI applications",

abstract = "The User Level Failure Mitigation (ULFM) interface has been proposed to provide fault-tolerant semantics in MPI. Previous work has presented performance evaluations of the interface; yet questions related to its programability and applicability remain unanswered. In this paper, we present our experiences on using ULFM in a case study (a large molecular dynamics application) to shed light on the advantages and difficulties of this interface to program fault-tolerant MPI applications. We found that, although ULFM is suitable for applications with work-decomposition flexibility (e.g., master-slave), it provides few benefits for more general (e.g., bulk synchronous) MPI applications.",

keywords = "Failure receovery models, Fault tolerance, MPI, Molecular dynamics simulation",

author = "Ignacio Laguna and Richards, {David F.} and Todd Gamblin and Martin Schulz and {De Supinski}, {Bronis R.}",

note = "Publisher Copyright: {\textcopyright} ACM 2014.; 21st European MPI Users' Group Meeting, EuroMPI/ASIA 2014 ; Conference date: 09-09-2014 Through 12-09-2014",

year = "2014",

month = sep,

day = "9",

doi = "10.1145/2642769.2642775",

language = "English",

series = "ACM International Conference Proceeding Series",

publisher = "Association for Computing Machinery",

pages = "57--62",

booktitle = "Proceedings of the 21st European MPI Users' Group Meeting, EuroMPI/ASIA 2014",

}

Laguna, I, Richards, DF, Gamblin, T, Schulz, M & De Supinski, BR 2014, Evaluating user-level fault tolerance for MPI applications. in Proceedings of the 21st European MPI Users' Group Meeting, EuroMPI/ASIA 2014. ACM International Conference Proceeding Series, vol. 09-12-September-2014, Association for Computing Machinery, pp. 57-62, 21st European MPI Users' Group Meeting, EuroMPI/ASIA 2014, Kyoto, Japan, 9/09/14. https://doi.org/10.1145/2642769.2642775

Evaluating user-level fault tolerance for MPI applications. / Laguna, Ignacio; Richards, David F.; Gamblin, Todd et al.
Proceedings of the 21st European MPI Users' Group Meeting, EuroMPI/ASIA 2014. Association for Computing Machinery, 2014. p. 57-62 (ACM International Conference Proceeding Series; Vol. 09-12-September-2014).

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

TY - GEN

T1 - Evaluating user-level fault tolerance for MPI applications

AU - Laguna, Ignacio

AU - Richards, David F.

AU - Gamblin, Todd

AU - Schulz, Martin

AU - De Supinski, Bronis R.

PY - 2014/9/9

Y1 - 2014/9/9

N2 - The User Level Failure Mitigation (ULFM) interface has been proposed to provide fault-tolerant semantics in MPI. Previous work has presented performance evaluations of the interface; yet questions related to its programability and applicability remain unanswered. In this paper, we present our experiences on using ULFM in a case study (a large molecular dynamics application) to shed light on the advantages and difficulties of this interface to program fault-tolerant MPI applications. We found that, although ULFM is suitable for applications with work-decomposition flexibility (e.g., master-slave), it provides few benefits for more general (e.g., bulk synchronous) MPI applications.

AB - The User Level Failure Mitigation (ULFM) interface has been proposed to provide fault-tolerant semantics in MPI. Previous work has presented performance evaluations of the interface; yet questions related to its programability and applicability remain unanswered. In this paper, we present our experiences on using ULFM in a case study (a large molecular dynamics application) to shed light on the advantages and difficulties of this interface to program fault-tolerant MPI applications. We found that, although ULFM is suitable for applications with work-decomposition flexibility (e.g., master-slave), it provides few benefits for more general (e.g., bulk synchronous) MPI applications.

KW - Failure receovery models

KW - Fault tolerance

KW - MPI

KW - Molecular dynamics simulation

UR - http://www.scopus.com/inward/record.url?scp=84958973677&partnerID=8YFLogxK

U2 - 10.1145/2642769.2642775

DO - 10.1145/2642769.2642775

M3 - Conference contribution

AN - SCOPUS:84958973677

T3 - ACM International Conference Proceeding Series

SP - 57

EP - 62

BT - Proceedings of the 21st European MPI Users' Group Meeting, EuroMPI/ASIA 2014

PB - Association for Computing Machinery

T2 - 21st European MPI Users' Group Meeting, EuroMPI/ASIA 2014

Y2 - 9 September 2014 through 12 September 2014

ER -

Evaluating user-level fault tolerance for MPI applications

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this