Supporting the development of resilient message passing applications using simulation

Thomas Naughton; Christian Engelmann; Geoffroy Vallee; Swen Bohm

doi:10.1109/PDP.2014.74

Supporting the development of resilient message passing applications using simulation

Thomas Naughton, Christian Engelmann, Geoffroy Vallee, Swen Bohm

Intelligent Systems and Facilities

Research output: Contribution to conference › Paper › peer-review

10 Scopus citations

Abstract

An emerging aspect of high-performance computing (HPC) hardware/software co-design is investigating performance under failure. The work in this paper extends the Extreme-scale Simulator (xSim), which was designed for evaluating the performance of message passing interface (MPI) applications on future HPC architectures, with fault-tolerant MPI extensions proposed by the MPI Fault Tolerance Working Group. xSim permits running MPI applications with millions of concurrent MPI ranks, while observing application performance in a simulated extreme-scale system using a lightweight parallel discrete event simulation. The newly added features offer user-level failure mitigation (ULFM) extensions at the simulated MPI layer to support algorithm-based fault tolerance (ABFT). The presented solution permits investigating performance under failure and failure handling of ABFT solutions. The newly enhanced xSim is the very first performance tool that supports ULFM and ABFT.

Original language	English
Pages	271-278
Number of pages	8
DOIs	https://doi.org/10.1109/PDP.2014.74
State	Published - 2014
Event	2014 22nd Euromicro International Conference on Parallel, Distributed, and Network-Based Processing, PDP 2014 - Turin, Italy Duration: Feb 12 2014 → Feb 14 2014

Conference

Conference	2014 22nd Euromicro International Conference on Parallel, Distributed, and Network-Based Processing, PDP 2014
Country/Territory	Italy
City	Turin
Period	02/12/14 → 02/14/14

Funding

Funders	Funder number
Oak Ridge National Laboratory

Keywords

Algorithm-based Fault Tolerance
High-performance Computing
Message Passing Interface
Parallel Discrete Event Simulation
Performance Prediction

Access to Document

10.1109/PDP.2014.74

Cite this

@conference{4c37fff1b1904c6e8242862b3aa2d24f,

title = "Supporting the development of resilient message passing applications using simulation",

abstract = "An emerging aspect of high-performance computing (HPC) hardware/software co-design is investigating performance under failure. The work in this paper extends the Extreme-scale Simulator (xSim), which was designed for evaluating the performance of message passing interface (MPI) applications on future HPC architectures, with fault-tolerant MPI extensions proposed by the MPI Fault Tolerance Working Group. xSim permits running MPI applications with millions of concurrent MPI ranks, while observing application performance in a simulated extreme-scale system using a lightweight parallel discrete event simulation. The newly added features offer user-level failure mitigation (ULFM) extensions at the simulated MPI layer to support algorithm-based fault tolerance (ABFT). The presented solution permits investigating performance under failure and failure handling of ABFT solutions. The newly enhanced xSim is the very first performance tool that supports ULFM and ABFT.",

keywords = "Algorithm-based Fault Tolerance, High-performance Computing, Message Passing Interface, Parallel Discrete Event Simulation, Performance Prediction",

author = "Thomas Naughton and Christian Engelmann and Geoffroy Vallee and Swen Bohm",

year = "2014",

doi = "10.1109/PDP.2014.74",

language = "English",

pages = "271--278",

note = "2014 22nd Euromicro International Conference on Parallel, Distributed, and Network-Based Processing, PDP 2014 ; Conference date: 12-02-2014 Through 14-02-2014",

}

Supporting the development of resilient message passing applications using simulation. / Naughton, Thomas ; Engelmann, Christian; Vallee, Geoffroy et al.
2014. 271-278 Paper presented at 2014 22nd Euromicro International Conference on Parallel, Distributed, and Network-Based Processing, PDP 2014, Turin, Italy.

Research output: Contribution to conference › Paper › peer-review

TY - CONF

T1 - Supporting the development of resilient message passing applications using simulation

AU - Naughton, Thomas

AU - Engelmann, Christian

AU - Vallee, Geoffroy

AU - Bohm, Swen

PY - 2014

Y1 - 2014

N2 - An emerging aspect of high-performance computing (HPC) hardware/software co-design is investigating performance under failure. The work in this paper extends the Extreme-scale Simulator (xSim), which was designed for evaluating the performance of message passing interface (MPI) applications on future HPC architectures, with fault-tolerant MPI extensions proposed by the MPI Fault Tolerance Working Group. xSim permits running MPI applications with millions of concurrent MPI ranks, while observing application performance in a simulated extreme-scale system using a lightweight parallel discrete event simulation. The newly added features offer user-level failure mitigation (ULFM) extensions at the simulated MPI layer to support algorithm-based fault tolerance (ABFT). The presented solution permits investigating performance under failure and failure handling of ABFT solutions. The newly enhanced xSim is the very first performance tool that supports ULFM and ABFT.

AB - An emerging aspect of high-performance computing (HPC) hardware/software co-design is investigating performance under failure. The work in this paper extends the Extreme-scale Simulator (xSim), which was designed for evaluating the performance of message passing interface (MPI) applications on future HPC architectures, with fault-tolerant MPI extensions proposed by the MPI Fault Tolerance Working Group. xSim permits running MPI applications with millions of concurrent MPI ranks, while observing application performance in a simulated extreme-scale system using a lightweight parallel discrete event simulation. The newly added features offer user-level failure mitigation (ULFM) extensions at the simulated MPI layer to support algorithm-based fault tolerance (ABFT). The presented solution permits investigating performance under failure and failure handling of ABFT solutions. The newly enhanced xSim is the very first performance tool that supports ULFM and ABFT.

KW - Algorithm-based Fault Tolerance

KW - High-performance Computing

KW - Message Passing Interface

KW - Parallel Discrete Event Simulation

KW - Performance Prediction

UR - http://www.scopus.com/inward/record.url?scp=84899463571&partnerID=8YFLogxK

U2 - 10.1109/PDP.2014.74

DO - 10.1109/PDP.2014.74

M3 - Paper

AN - SCOPUS:84899463571

SP - 271

EP - 278

T2 - 2014 22nd Euromicro International Conference on Parallel, Distributed, and Network-Based Processing, PDP 2014

Y2 - 12 February 2014 through 14 February 2014

ER -

Supporting the development of resilient message passing applications using simulation

Abstract

Conference

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this