Balancing performance and portability with containers in HPC: An OpenSHMEM example

Thomas Naughton; Lawrence Sorrillo; Adam Simpson; Neena Imam

doi:10.1007/978-3-319-73814-7_9

Balancing performance and portability with containers in HPC: An OpenSHMEM example

Thomas Naughton, Lawrence Sorrillo, Adam Simpson, Neena Imam

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

2 Scopus citations

Abstract

There is a growing interest in using Linux containers to streamline software development and application deployment. A container enables the user to bundle the salient elements of the software stack from an application’s perspective. In this paper, we discuss initial experiences in using the Open MPI implementation of OpenSHMEM with containers on HPC resources. We provide a brief overview of two container runtimes, Docker & Singularity, highlighting elements that are of interest for HPC users. The Docker platform offers a rich set of services that are widely used in enterprise environments, whereas Singularity is an emerging container runtime that is specifically written for use on HPC systems. We describe our procedure for container assembly and deployment that strives to maintain the portability of the container-based application. We show performance results for the Graph500 benchmark running along the typical continuum of development testbed up to full production supercomputer (ORNL’s Titan). The results show consistent performance between the native and Singularity (container) tests. The results also showed an unexplained drop in performance when using the Cray Gemini network with Open MPI’s OpenSHMEM, which was unrelated to the container usage.

Original language	English
Title of host publication	OpenSHMEM and Related Technologies
Subtitle of host publication	Big Compute and Big Data Convergence - 4th Workshop, OpenSHMEM 2017, Revised Selected Papers
Editors	Manjunath Gorentla Venkata, Neena Imam, Swaroop Pophale
Publisher	Springer Verlag
Pages	130-142
Number of pages	13
ISBN (Print)	9783319738130
DOIs	https://doi.org/10.1007/978-3-319-73814-7_9
State	Published - 2018
Event	4th Workshop on OpenSHMEM and Related Technologies: Big Compute and Big Data Convergence, OpenSHMEM 2017 - Annapolis, United States Duration: Aug 7 2017 → Aug 9 2017

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	10679 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	4th Workshop on OpenSHMEM and Related Technologies: Big Compute and Big Data Convergence, OpenSHMEM 2017
Country/Territory	United States
City	Annapolis
Period	08/7/17 → 08/9/17

Funding

This work was sponsored by the U.S. Department of Energy’s Office of Advanced Scientific Computing Research. This manuscript has been authored by UT-Battelle, LLC under Contract No. DE-AC05-00OR22725 with the U.S. Department of Energy. The United States Government retains and the publisher, by accepting the article for publication, acknowledges that the United States Government retains a nonexclusive, paid-up, irrevocable, world-wide license to publish or reproduce the published form of this manuscript, or allow others to do so, for United States Government purposes. The Department of Energy will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan (http:// energy.gov/downloads/doe-public-access-plan). This research used resources of the Oak Ridge Leadership Computing Facility, which is a DOE Office of Science User Facility supported under Contract DE-AC05-00OR22725.

Funders	Funder number
DOE Office of Science
U.S. Department of Energy	DE-AC05-00OR22725
Advanced Scientific Computing Research

Access to Document

10.1007/978-3-319-73814-7_9

Cite this

Naughton, T., Sorrillo, L., Simpson, A., & Imam, N. (2018). Balancing performance and portability with containers in HPC: An OpenSHMEM example. In M. Gorentla Venkata, N. Imam, & S. Pophale (Eds.), OpenSHMEM and Related Technologies: Big Compute and Big Data Convergence - 4th Workshop, OpenSHMEM 2017, Revised Selected Papers (pp. 130-142). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10679 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-73814-7_9

Naughton, Thomas ; Sorrillo, Lawrence ; Simpson, Adam et al. / Balancing performance and portability with containers in HPC : An OpenSHMEM example. OpenSHMEM and Related Technologies: Big Compute and Big Data Convergence - 4th Workshop, OpenSHMEM 2017, Revised Selected Papers. editor / Manjunath Gorentla Venkata ; Neena Imam ; Swaroop Pophale. Springer Verlag, 2018. pp. 130-142 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{f14d83eda3ae4c91a468646eeefcd798,

title = "Balancing performance and portability with containers in HPC: An OpenSHMEM example",

abstract = "There is a growing interest in using Linux containers to streamline software development and application deployment. A container enables the user to bundle the salient elements of the software stack from an application{\textquoteright}s perspective. In this paper, we discuss initial experiences in using the Open MPI implementation of OpenSHMEM with containers on HPC resources. We provide a brief overview of two container runtimes, Docker & Singularity, highlighting elements that are of interest for HPC users. The Docker platform offers a rich set of services that are widely used in enterprise environments, whereas Singularity is an emerging container runtime that is specifically written for use on HPC systems. We describe our procedure for container assembly and deployment that strives to maintain the portability of the container-based application. We show performance results for the Graph500 benchmark running along the typical continuum of development testbed up to full production supercomputer (ORNL{\textquoteright}s Titan). The results show consistent performance between the native and Singularity (container) tests. The results also showed an unexplained drop in performance when using the Cray Gemini network with Open MPI{\textquoteright}s OpenSHMEM, which was unrelated to the container usage.",

author = "Thomas Naughton and Lawrence Sorrillo and Adam Simpson and Neena Imam",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing AG 2018.; 4th Workshop on OpenSHMEM and Related Technologies: Big Compute and Big Data Convergence, OpenSHMEM 2017 ; Conference date: 07-08-2017 Through 09-08-2017",

year = "2018",

doi = "10.1007/978-3-319-73814-7_9",

language = "English",

isbn = "9783319738130",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "130--142",

editor = "{Gorentla Venkata}, Manjunath and Neena Imam and Swaroop Pophale",

booktitle = "OpenSHMEM and Related Technologies",

}

Naughton, T , Sorrillo, L, Simpson, A & Imam, N 2018, Balancing performance and portability with containers in HPC: An OpenSHMEM example. in M Gorentla Venkata, N Imam & S Pophale (eds), OpenSHMEM and Related Technologies: Big Compute and Big Data Convergence - 4th Workshop, OpenSHMEM 2017, Revised Selected Papers. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10679 LNCS, Springer Verlag, pp. 130-142, 4th Workshop on OpenSHMEM and Related Technologies: Big Compute and Big Data Convergence, OpenSHMEM 2017, Annapolis, United States, 08/7/17. https://doi.org/10.1007/978-3-319-73814-7_9

Balancing performance and portability with containers in HPC: An OpenSHMEM example. / Naughton, Thomas ; Sorrillo, Lawrence; Simpson, Adam et al.
OpenSHMEM and Related Technologies: Big Compute and Big Data Convergence - 4th Workshop, OpenSHMEM 2017, Revised Selected Papers. ed. / Manjunath Gorentla Venkata; Neena Imam; Swaroop Pophale. Springer Verlag, 2018. p. 130-142 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10679 LNCS).

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

TY - GEN

T1 - Balancing performance and portability with containers in HPC

T2 - 4th Workshop on OpenSHMEM and Related Technologies: Big Compute and Big Data Convergence, OpenSHMEM 2017

AU - Naughton, Thomas

AU - Sorrillo, Lawrence

AU - Simpson, Adam

AU - Imam, Neena

PY - 2018

Y1 - 2018

N2 - There is a growing interest in using Linux containers to streamline software development and application deployment. A container enables the user to bundle the salient elements of the software stack from an application’s perspective. In this paper, we discuss initial experiences in using the Open MPI implementation of OpenSHMEM with containers on HPC resources. We provide a brief overview of two container runtimes, Docker & Singularity, highlighting elements that are of interest for HPC users. The Docker platform offers a rich set of services that are widely used in enterprise environments, whereas Singularity is an emerging container runtime that is specifically written for use on HPC systems. We describe our procedure for container assembly and deployment that strives to maintain the portability of the container-based application. We show performance results for the Graph500 benchmark running along the typical continuum of development testbed up to full production supercomputer (ORNL’s Titan). The results show consistent performance between the native and Singularity (container) tests. The results also showed an unexplained drop in performance when using the Cray Gemini network with Open MPI’s OpenSHMEM, which was unrelated to the container usage.

AB - There is a growing interest in using Linux containers to streamline software development and application deployment. A container enables the user to bundle the salient elements of the software stack from an application’s perspective. In this paper, we discuss initial experiences in using the Open MPI implementation of OpenSHMEM with containers on HPC resources. We provide a brief overview of two container runtimes, Docker & Singularity, highlighting elements that are of interest for HPC users. The Docker platform offers a rich set of services that are widely used in enterprise environments, whereas Singularity is an emerging container runtime that is specifically written for use on HPC systems. We describe our procedure for container assembly and deployment that strives to maintain the portability of the container-based application. We show performance results for the Graph500 benchmark running along the typical continuum of development testbed up to full production supercomputer (ORNL’s Titan). The results show consistent performance between the native and Singularity (container) tests. The results also showed an unexplained drop in performance when using the Cray Gemini network with Open MPI’s OpenSHMEM, which was unrelated to the container usage.

UR - http://www.scopus.com/inward/record.url?scp=85041539078&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-73814-7_9

DO - 10.1007/978-3-319-73814-7_9

M3 - Conference contribution

AN - SCOPUS:85041539078

SN - 9783319738130

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 130

EP - 142

BT - OpenSHMEM and Related Technologies

A2 - Gorentla Venkata, Manjunath

A2 - Imam, Neena

A2 - Pophale, Swaroop

PB - Springer Verlag

Y2 - 7 August 2017 through 9 August 2017

ER -

Naughton T , Sorrillo L, Simpson A, Imam N. Balancing performance and portability with containers in HPC: An OpenSHMEM example. In Gorentla Venkata M, Imam N, Pophale S, editors, OpenSHMEM and Related Technologies: Big Compute and Big Data Convergence - 4th Workshop, OpenSHMEM 2017, Revised Selected Papers. Springer Verlag. 2018. p. 130-142. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-73814-7_9

Balancing performance and portability with containers in HPC: An OpenSHMEM example

Abstract

Publication series

Conference

Funding

Access to Document

Other files and links

Fingerprint

Cite this