Preparation and optimization of a diverse workload for a large-scale heterogeneous system

Ian Karlin; Yoonho Park; Bronis R. De Supinski; Peng Wang; Bert Still; David Beckingsale; Robert Blake; Tong Chen; Guojing Cong; Carlos Costa; Johann Dahm; Giacomo Domeniconi; Thomas Epperly; Aaron Fisher; Sara Kokkila Schumacher; Steven Langer; Hai Le; Eun Kyung Lee; Naoya Maruyama; Xinyu Que; David Richards; Bjorn Sjogreen; Jonathan Wong; Carol Woodward; Ulrike Yang; Xiaohua Zhang; Bob Anderson; David Appelhans; Levi Barnes; Peter Barnes; Sorin Bastea; David Boehme; Jamie A. Bramwell; Jim Brase; Jose Brunheroto; Barry Chen; Charway R. Cooper; Tony Degroot; Rob Falgout; Todd Gamblin; David Gardner; James Glosli; John Gunnels; Max Katz; Tzanio Kolev; I. Feng W. Kuo; Matthew P. Legendre; Ruipeng Li; Pei Hung Lin; Shelby Lockhart; Kathleen McCandless; Claudia Misale; Jaime Moreno; Rob Neely; Jarom Nelson; Rao Nimmakayala; Kathryn O'Brien; Kevin O'Brien; Ramesh Pankajakshan; Roger Pearce; Slaven Peles; Phil Regier; Steve Rennich; Martin Schulz; Howard Scott; James Sexton; Kathleen Shoga; Shiv Sundram; G. Thomas-Collignon; Brian Van Essen; Alexey Voronin; Bob Walkup; Lu Wang; Chris Ward; Hui Fang Wen; Dan White; Christopher Young; Cyril Zeller; Ed Zywicz

doi:10.1145/3295500.3356192

Preparation and optimization of a diverse workload for a large-scale heterogeneous system

Ian Karlin, Yoonho Park, Bronis R. De Supinski, Peng Wang, Bert Still, David Beckingsale, Robert Blake, Tong Chen, Guojing Cong, Carlos Costa, Johann Dahm, Giacomo Domeniconi, Thomas Epperly, Aaron Fisher, Sara Kokkila Schumacher, Steven Langer, Hai Le, Eun Kyung Lee, Naoya Maruyama, Xinyu QueDavid Richards, Bjorn Sjogreen, Jonathan Wong, Carol Woodward, Ulrike Yang, Xiaohua Zhang, Bob Anderson, David Appelhans, Levi Barnes, Peter Barnes, Sorin Bastea, David Boehme, Jamie A. Bramwell, Jim Brase, Jose Brunheroto, Barry Chen, Charway R. Cooper, Tony Degroot, Rob Falgout, Todd Gamblin, David Gardner, James Glosli, John Gunnels, Max Katz, Tzanio Kolev, I. Feng W. Kuo, Matthew P. Legendre, Ruipeng Li, Pei Hung Lin, Shelby Lockhart, Kathleen McCandless, Claudia Misale, Jaime Moreno, Rob Neely, Jarom Nelson, Rao Nimmakayala, Kathryn O'Brien, Kevin O'Brien, Ramesh Pankajakshan, Roger Pearce, Slaven Peles, Phil Regier, Steve Rennich, Martin Schulz, Howard Scott, James Sexton, Kathleen Shoga, Shiv Sundram, G. Thomas-Collignon, Brian Van Essen, Alexey Voronin, Bob Walkup, Lu Wang, Chris Ward, Hui Fang Wen, Dan White, Christopher Young, Cyril Zeller, Ed Zywicz

Learning Systems

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

3 Scopus citations

Abstract

Productivity from day one on supercomputers that leverage new technologies requires significant preparation. An institution that procures a novel system architecture often lacks sufficient institutional knowledge and skills to prepare for it. Thus, the "Center of Excellence" (CoE) concept has emerged to prepare for systems such as Summit and Sierra, currently the top two systems in the Top 500. This paper documents CoE experiences that prepared a workload of diverse applications and math libraries for a heterogeneous system. We describe our approach to this preparation, including our management and execution strategies, and detail our experiences with and reasons for using different programming approaches. Our early science and performance results show that the project enabled significant early seismic science with up to a l4X throughput increase over Cori. In addition to our successes, we discuss our challenges and failures so others may benefit from our experience.

Original language	English
Title of host publication	Proceedings of SC 2019
Subtitle of host publication	The International Conference for High Performance Computing, Networking, Storage and Analysis
Publisher	IEEE Computer Society
ISBN (Electronic)	9781450362290
DOIs	https://doi.org/10.1145/3295500.3356192
State	Published - Nov 17 2019
Event	2019 International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2019 - Denver, United States Duration: Nov 17 2019 → Nov 22 2019

Publication series

Name	International Conference for High Performance Computing, Networking, Storage and Analysis, SC
ISSN (Print)	2167-4329
ISSN (Electronic)	2167-4337

Conference

Conference	2019 International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2019
Country/Territory	United States
City	Denver
Period	11/17/19 → 11/22/19

Funding

Prepared by LLNL under Contract DE-AC52-07NA27344. LLNL-CONF-772139. IBM and NVIDIA participation was supported under CORAL NRE Contract B604142.

Keywords

GPUs
Heterogeneous systems
Large-scale applications
Performance
Project management
programming models

Access to Document

10.1145/3295500.3356192

Cite this

Karlin, I., Park, Y., De Supinski, B. R., Wang, P., Still, B., Beckingsale, D., Blake, R., Chen, T., Cong, G., Costa, C., Dahm, J., Domeniconi, G., Epperly, T., Fisher, A., Schumacher, S. K., Langer, S., Le, H., Lee, E. K., Maruyama, N., ... Zywicz, E. (2019). Preparation and optimization of a diverse workload for a large-scale heterogeneous system. In Proceedings of SC 2019: The International Conference for High Performance Computing, Networking, Storage and Analysis Article a32 (International Conference for High Performance Computing, Networking, Storage and Analysis, SC). IEEE Computer Society. https://doi.org/10.1145/3295500.3356192

Karlin, Ian ; Park, Yoonho ; De Supinski, Bronis R. et al. / Preparation and optimization of a diverse workload for a large-scale heterogeneous system. Proceedings of SC 2019: The International Conference for High Performance Computing, Networking, Storage and Analysis. IEEE Computer Society, 2019. (International Conference for High Performance Computing, Networking, Storage and Analysis, SC).

@inproceedings{1ea9a25f695f460ea05c35ab341d7c95,

title = "Preparation and optimization of a diverse workload for a large-scale heterogeneous system",

abstract = "Productivity from day one on supercomputers that leverage new technologies requires significant preparation. An institution that procures a novel system architecture often lacks sufficient institutional knowledge and skills to prepare for it. Thus, the {"}Center of Excellence{"} (CoE) concept has emerged to prepare for systems such as Summit and Sierra, currently the top two systems in the Top 500. This paper documents CoE experiences that prepared a workload of diverse applications and math libraries for a heterogeneous system. We describe our approach to this preparation, including our management and execution strategies, and detail our experiences with and reasons for using different programming approaches. Our early science and performance results show that the project enabled significant early seismic science with up to a l4X throughput increase over Cori. In addition to our successes, we discuss our challenges and failures so others may benefit from our experience.",

keywords = "GPUs, Heterogeneous systems, Large-scale applications, Performance, Project management, programming models",

author = "Ian Karlin and Yoonho Park and \{De Supinski\}, \{Bronis R.\} and Peng Wang and Bert Still and David Beckingsale and Robert Blake and Tong Chen and Guojing Cong and Carlos Costa and Johann Dahm and Giacomo Domeniconi and Thomas Epperly and Aaron Fisher and Schumacher, \{Sara Kokkila\} and Steven Langer and Hai Le and Lee, \{Eun Kyung\} and Naoya Maruyama and Xinyu Que and David Richards and Bjorn Sjogreen and Jonathan Wong and Carol Woodward and Ulrike Yang and Xiaohua Zhang and Bob Anderson and David Appelhans and Levi Barnes and Peter Barnes and Sorin Bastea and David Boehme and Bramwell, \{Jamie A.\} and Jim Brase and Jose Brunheroto and Barry Chen and Cooper, \{Charway R.\} and Tony Degroot and Rob Falgout and Todd Gamblin and David Gardner and James Glosli and John Gunnels and Max Katz and Tzanio Kolev and Kuo, \{I. Feng W.\} and Legendre, \{Matthew P.\} and Ruipeng Li and Lin, \{Pei Hung\} and Shelby Lockhart and Kathleen McCandless and Claudia Misale and Jaime Moreno and Rob Neely and Jarom Nelson and Rao Nimmakayala and Kathryn O'Brien and Kevin O'Brien and Ramesh Pankajakshan and Roger Pearce and Slaven Peles and Phil Regier and Steve Rennich and Martin Schulz and Howard Scott and James Sexton and Kathleen Shoga and Shiv Sundram and G. Thomas-Collignon and \{Van Essen\}, Brian and Alexey Voronin and Bob Walkup and Lu Wang and Chris Ward and Wen, \{Hui Fang\} and Dan White and Christopher Young and Cyril Zeller and Ed Zywicz",

note = "Publisher Copyright: {\textcopyright} 2019 ACM.; 2019 International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2019 ; Conference date: 17-11-2019 Through 22-11-2019",

year = "2019",

month = nov,

day = "17",

doi = "10.1145/3295500.3356192",

language = "English",

series = "International Conference for High Performance Computing, Networking, Storage and Analysis, SC",

publisher = "IEEE Computer Society",

booktitle = "Proceedings of SC 2019",

}

Karlin, I, Park, Y, De Supinski, BR, Wang, P, Still, B, Beckingsale, D, Blake, R, Chen, T, Cong, G, Costa, C, Dahm, J, Domeniconi, G, Epperly, T, Fisher, A, Schumacher, SK, Langer, S, Le, H, Lee, EK, Maruyama, N, Que, X, Richards, D, Sjogreen, B, Wong, J, Woodward, C, Yang, U, Zhang, X, Anderson, B, Appelhans, D, Barnes, L, Barnes, P, Bastea, S, Boehme, D, Bramwell, JA, Brase, J, Brunheroto, J, Chen, B, Cooper, CR, Degroot, T, Falgout, R, Gamblin, T, Gardner, D, Glosli, J, Gunnels, J, Katz, M, Kolev, T, Kuo, IFW, Legendre, MP, Li, R, Lin, PH, Lockhart, S, McCandless, K, Misale, C, Moreno, J, Neely, R, Nelson, J, Nimmakayala, R, O'Brien, K, O'Brien, K, Pankajakshan, R, Pearce, R, Peles, S, Regier, P, Rennich, S, Schulz, M, Scott, H, Sexton, J, Shoga, K, Sundram, S, Thomas-Collignon, G, Van Essen, B, Voronin, A, Walkup, B, Wang, L, Ward, C, Wen, HF, White, D, Young, C, Zeller, C & Zywicz, E 2019, Preparation and optimization of a diverse workload for a large-scale heterogeneous system. in Proceedings of SC 2019: The International Conference for High Performance Computing, Networking, Storage and Analysis., a32, International Conference for High Performance Computing, Networking, Storage and Analysis, SC, IEEE Computer Society, 2019 International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2019, Denver, United States, 11/17/19. https://doi.org/10.1145/3295500.3356192

Preparation and optimization of a diverse workload for a large-scale heterogeneous system. / Karlin, Ian; Park, Yoonho; De Supinski, Bronis R. et al.
Proceedings of SC 2019: The International Conference for High Performance Computing, Networking, Storage and Analysis. IEEE Computer Society, 2019. a32 (International Conference for High Performance Computing, Networking, Storage and Analysis, SC).

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

TY - GEN

T1 - Preparation and optimization of a diverse workload for a large-scale heterogeneous system

AU - Karlin, Ian

AU - Park, Yoonho

AU - De Supinski, Bronis R.

AU - Wang, Peng

AU - Still, Bert

AU - Beckingsale, David

AU - Blake, Robert

AU - Chen, Tong

AU - Cong, Guojing

AU - Costa, Carlos

AU - Dahm, Johann

AU - Domeniconi, Giacomo

AU - Epperly, Thomas

AU - Fisher, Aaron

AU - Schumacher, Sara Kokkila

AU - Langer, Steven

AU - Le, Hai

AU - Lee, Eun Kyung

AU - Maruyama, Naoya

AU - Que, Xinyu

AU - Richards, David

AU - Sjogreen, Bjorn

AU - Wong, Jonathan

AU - Woodward, Carol

AU - Yang, Ulrike

AU - Zhang, Xiaohua

AU - Anderson, Bob

AU - Appelhans, David

AU - Barnes, Levi

AU - Barnes, Peter

AU - Bastea, Sorin

AU - Boehme, David

AU - Bramwell, Jamie A.

AU - Brase, Jim

AU - Brunheroto, Jose

AU - Chen, Barry

AU - Cooper, Charway R.

AU - Degroot, Tony

AU - Falgout, Rob

AU - Gamblin, Todd

AU - Gardner, David

AU - Glosli, James

AU - Gunnels, John

AU - Katz, Max

AU - Kolev, Tzanio

AU - Kuo, I. Feng W.

AU - Legendre, Matthew P.

AU - Li, Ruipeng

AU - Lin, Pei Hung

AU - Lockhart, Shelby

AU - McCandless, Kathleen

AU - Misale, Claudia

AU - Moreno, Jaime

AU - Neely, Rob

AU - Nelson, Jarom

AU - Nimmakayala, Rao

AU - O'Brien, Kathryn

AU - O'Brien, Kevin

AU - Pankajakshan, Ramesh

AU - Pearce, Roger

AU - Peles, Slaven

AU - Regier, Phil

AU - Rennich, Steve

AU - Schulz, Martin

AU - Scott, Howard

AU - Sexton, James

AU - Shoga, Kathleen

AU - Sundram, Shiv

AU - Thomas-Collignon, G.

AU - Van Essen, Brian

AU - Voronin, Alexey

AU - Walkup, Bob

AU - Wang, Lu

AU - Ward, Chris

AU - Wen, Hui Fang

AU - White, Dan

AU - Young, Christopher

AU - Zeller, Cyril

AU - Zywicz, Ed

PY - 2019/11/17

Y1 - 2019/11/17

N2 - Productivity from day one on supercomputers that leverage new technologies requires significant preparation. An institution that procures a novel system architecture often lacks sufficient institutional knowledge and skills to prepare for it. Thus, the "Center of Excellence" (CoE) concept has emerged to prepare for systems such as Summit and Sierra, currently the top two systems in the Top 500. This paper documents CoE experiences that prepared a workload of diverse applications and math libraries for a heterogeneous system. We describe our approach to this preparation, including our management and execution strategies, and detail our experiences with and reasons for using different programming approaches. Our early science and performance results show that the project enabled significant early seismic science with up to a l4X throughput increase over Cori. In addition to our successes, we discuss our challenges and failures so others may benefit from our experience.

AB - Productivity from day one on supercomputers that leverage new technologies requires significant preparation. An institution that procures a novel system architecture often lacks sufficient institutional knowledge and skills to prepare for it. Thus, the "Center of Excellence" (CoE) concept has emerged to prepare for systems such as Summit and Sierra, currently the top two systems in the Top 500. This paper documents CoE experiences that prepared a workload of diverse applications and math libraries for a heterogeneous system. We describe our approach to this preparation, including our management and execution strategies, and detail our experiences with and reasons for using different programming approaches. Our early science and performance results show that the project enabled significant early seismic science with up to a l4X throughput increase over Cori. In addition to our successes, we discuss our challenges and failures so others may benefit from our experience.

KW - GPUs

KW - Heterogeneous systems

KW - Large-scale applications

KW - Performance

KW - Project management

KW - programming models

UR - https://www.scopus.com/pages/publications/85076255109

U2 - 10.1145/3295500.3356192

DO - 10.1145/3295500.3356192

M3 - Conference contribution

AN - SCOPUS:85076255109

T3 - International Conference for High Performance Computing, Networking, Storage and Analysis, SC

BT - Proceedings of SC 2019

PB - IEEE Computer Society

T2 - 2019 International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2019

Y2 - 17 November 2019 through 22 November 2019

ER -

Karlin I, Park Y, De Supinski BR, Wang P, Still B, Beckingsale D et al. Preparation and optimization of a diverse workload for a large-scale heterogeneous system. In Proceedings of SC 2019: The International Conference for High Performance Computing, Networking, Storage and Analysis. IEEE Computer Society. 2019. a32. (International Conference for High Performance Computing, Networking, Storage and Analysis, SC). doi: 10.1145/3295500.3356192

Preparation and optimization of a diverse workload for a large-scale heterogeneous system

Abstract

Publication series

Conference

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this