Performance scaling variability and energy analysis for a resilient ULFM-based PDE solver

K. Morris; F. Rizzi; B. Cook; P. Mycek; O. LeMaitre; O. M. Knio; K. Sargsyan; K. Dahlgren; B. J. Debusschere

doi:10.1109/ScalA.2016.010

Performance scaling variability and energy analysis for a resilient ULFM-based PDE solver

K. Morris, F. Rizzi, B. Cook, P. Mycek, O. LeMaitre, O. M. Knio, K. Sargsyan, K. Dahlgren, B. J. Debusschere

Applied Mathematics and Computational Science

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

1 Scopus citations

Abstract

We present a resilient task-based domain-decomposition preconditioner for partial differential equations (PDEs) built on top of User Level Fault Mitigation Message Passing Interface (ULFM-MPI). The algorithm reformulates the PDE as a sampling problem, followed by a robust regression-based solution update that is resilient to silent data corruptions (SDCs). We adopt a server-client model where all state information is held by the servers, while clients only serve as computational units. The task-based nature of the algorithm and the capabilities of ULFM complement each other to support missing tasks, making the application resilient to clients failing.We present weak and strong scaling results on Edison, National Energy Research Scientific Computing Center (NERSC), for a nominal and a fault-injected case, showing that even in the presence of faults, scalability tested up to 50k cores is within 90%. We then quantify the variability of weak and strong scaling due to the presence of faults. Finally, we discuss the performance of our application with respect to subdomain size, server/client configuration, and the interplay between energy and resilience.

Original language	English (US)
Title of host publication	Proceedings of ScalA 2016
Subtitle of host publication	7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems - Held in conjunction with SC16: The International Conference for High Performance Computing, Networking, Storage and Analysis
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	41-48
Number of pages	8
ISBN (Electronic)	9781509052226
DOIs	https://doi.org/10.1109/ScalA.2016.010
State	Published - Jan 30 2017
Event	7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems, ScalA 2016 - Salt Lake City, United States Duration: Nov 13 2016 → Nov 18 2016

Publication series

Name	Proceedings of ScalA 2016: 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems - Held in conjunction with SC16: The International Conference for High Performance Computing, Networking, Storage and Analysis

Other

Other	7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems, ScalA 2016
Country/Territory	United States
City	Salt Lake City
Period	11/13/16 → 11/18/16

Bibliographical note

Publisher Copyright:
© 2016 IEEE.

Keywords

Client-server systems
Dynamic voltage scaling
Fault tolerance
Partial differential equations
Resilience

ASJC Scopus subject areas

Computer Science Applications
Numerical Analysis
Software
Computational Mathematics

Access to Document

10.1109/ScalA.2016.010

Cite this

Morris, K., Rizzi, F., Cook, B., Mycek, P., LeMaitre, O., Knio, O. M., Sargsyan, K., Dahlgren, K., & Debusschere, B. J. (2017). Performance scaling variability and energy analysis for a resilient ULFM-based PDE solver. In Proceedings of ScalA 2016: 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems - Held in conjunction with SC16: The International Conference for High Performance Computing, Networking, Storage and Analysis (pp. 41-48). Article 7836595 (Proceedings of ScalA 2016: 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems - Held in conjunction with SC16: The International Conference for High Performance Computing, Networking, Storage and Analysis). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ScalA.2016.010

Morris, K. ; Rizzi, F. ; Cook, B. et al. / Performance scaling variability and energy analysis for a resilient ULFM-based PDE solver. Proceedings of ScalA 2016: 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems - Held in conjunction with SC16: The International Conference for High Performance Computing, Networking, Storage and Analysis. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 41-48 (Proceedings of ScalA 2016: 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems - Held in conjunction with SC16: The International Conference for High Performance Computing, Networking, Storage and Analysis).

@inproceedings{cab2bdef98a84392bd2eac1caa6dac0f,

title = "Performance scaling variability and energy analysis for a resilient ULFM-based PDE solver",

abstract = "We present a resilient task-based domain-decomposition preconditioner for partial differential equations (PDEs) built on top of User Level Fault Mitigation Message Passing Interface (ULFM-MPI). The algorithm reformulates the PDE as a sampling problem, followed by a robust regression-based solution update that is resilient to silent data corruptions (SDCs). We adopt a server-client model where all state information is held by the servers, while clients only serve as computational units. The task-based nature of the algorithm and the capabilities of ULFM complement each other to support missing tasks, making the application resilient to clients failing.We present weak and strong scaling results on Edison, National Energy Research Scientific Computing Center (NERSC), for a nominal and a fault-injected case, showing that even in the presence of faults, scalability tested up to 50k cores is within 90%. We then quantify the variability of weak and strong scaling due to the presence of faults. Finally, we discuss the performance of our application with respect to subdomain size, server/client configuration, and the interplay between energy and resilience.",

keywords = "Client-server systems, Dynamic voltage scaling, Fault tolerance, Partial differential equations, Resilience",

author = "K. Morris and F. Rizzi and B. Cook and P. Mycek and O. LeMaitre and Knio, {O. M.} and K. Sargsyan and K. Dahlgren and Debusschere, {B. J.}",

note = "Publisher Copyright: {\textcopyright} 2016 IEEE.; 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems, ScalA 2016 ; Conference date: 13-11-2016 Through 18-11-2016",

year = "2017",

month = jan,

day = "30",

doi = "10.1109/ScalA.2016.010",

language = "English (US)",

series = "Proceedings of ScalA 2016: 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems - Held in conjunction with SC16: The International Conference for High Performance Computing, Networking, Storage and Analysis",

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

pages = "41--48",

booktitle = "Proceedings of ScalA 2016",

address = "United States",

}

Morris, K, Rizzi, F, Cook, B, Mycek, P, LeMaitre, O, Knio, OM, Sargsyan, K, Dahlgren, K & Debusschere, BJ 2017, Performance scaling variability and energy analysis for a resilient ULFM-based PDE solver. in Proceedings of ScalA 2016: 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems - Held in conjunction with SC16: The International Conference for High Performance Computing, Networking, Storage and Analysis., 7836595, Proceedings of ScalA 2016: 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems - Held in conjunction with SC16: The International Conference for High Performance Computing, Networking, Storage and Analysis, Institute of Electrical and Electronics Engineers Inc., pp. 41-48, 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems, ScalA 2016, Salt Lake City, United States, 11/13/16. https://doi.org/10.1109/ScalA.2016.010

Performance scaling variability and energy analysis for a resilient ULFM-based PDE solver. / Morris, K.; Rizzi, F.; Cook, B. et al.
Proceedings of ScalA 2016: 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems - Held in conjunction with SC16: The International Conference for High Performance Computing, Networking, Storage and Analysis. Institute of Electrical and Electronics Engineers Inc., 2017. p. 41-48 7836595 (Proceedings of ScalA 2016: 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems - Held in conjunction with SC16: The International Conference for High Performance Computing, Networking, Storage and Analysis).

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

TY - GEN

T1 - Performance scaling variability and energy analysis for a resilient ULFM-based PDE solver

AU - Morris, K.

AU - Rizzi, F.

AU - Cook, B.

AU - Mycek, P.

AU - LeMaitre, O.

AU - Knio, O. M.

AU - Sargsyan, K.

AU - Dahlgren, K.

AU - Debusschere, B. J.

PY - 2017/1/30

Y1 - 2017/1/30

N2 - We present a resilient task-based domain-decomposition preconditioner for partial differential equations (PDEs) built on top of User Level Fault Mitigation Message Passing Interface (ULFM-MPI). The algorithm reformulates the PDE as a sampling problem, followed by a robust regression-based solution update that is resilient to silent data corruptions (SDCs). We adopt a server-client model where all state information is held by the servers, while clients only serve as computational units. The task-based nature of the algorithm and the capabilities of ULFM complement each other to support missing tasks, making the application resilient to clients failing.We present weak and strong scaling results on Edison, National Energy Research Scientific Computing Center (NERSC), for a nominal and a fault-injected case, showing that even in the presence of faults, scalability tested up to 50k cores is within 90%. We then quantify the variability of weak and strong scaling due to the presence of faults. Finally, we discuss the performance of our application with respect to subdomain size, server/client configuration, and the interplay between energy and resilience.

AB - We present a resilient task-based domain-decomposition preconditioner for partial differential equations (PDEs) built on top of User Level Fault Mitigation Message Passing Interface (ULFM-MPI). The algorithm reformulates the PDE as a sampling problem, followed by a robust regression-based solution update that is resilient to silent data corruptions (SDCs). We adopt a server-client model where all state information is held by the servers, while clients only serve as computational units. The task-based nature of the algorithm and the capabilities of ULFM complement each other to support missing tasks, making the application resilient to clients failing.We present weak and strong scaling results on Edison, National Energy Research Scientific Computing Center (NERSC), for a nominal and a fault-injected case, showing that even in the presence of faults, scalability tested up to 50k cores is within 90%. We then quantify the variability of weak and strong scaling due to the presence of faults. Finally, we discuss the performance of our application with respect to subdomain size, server/client configuration, and the interplay between energy and resilience.

KW - Client-server systems

KW - Dynamic voltage scaling

KW - Fault tolerance

KW - Partial differential equations

KW - Resilience

UR - http://www.scopus.com/inward/record.url?scp=85015189347&partnerID=8YFLogxK

U2 - 10.1109/ScalA.2016.010

DO - 10.1109/ScalA.2016.010

M3 - Conference contribution

AN - SCOPUS:85015189347

T3 - Proceedings of ScalA 2016: 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems - Held in conjunction with SC16: The International Conference for High Performance Computing, Networking, Storage and Analysis

SP - 41

EP - 48

BT - Proceedings of ScalA 2016

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems, ScalA 2016

Y2 - 13 November 2016 through 18 November 2016

ER -

Morris K, Rizzi F, Cook B, Mycek P, LeMaitre O, Knio OM et al. Performance scaling variability and energy analysis for a resilient ULFM-based PDE solver. In Proceedings of ScalA 2016: 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems - Held in conjunction with SC16: The International Conference for High Performance Computing, Networking, Storage and Analysis. Institute of Electrical and Electronics Engineers Inc. 2017. p. 41-48. 7836595. (Proceedings of ScalA 2016: 7th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems - Held in conjunction with SC16: The International Conference for High Performance Computing, Networking, Storage and Analysis). doi: 10.1109/ScalA.2016.010

Performance scaling variability and energy analysis for a resilient ULFM-based PDE solver

Abstract

Publication series

Other

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this