Performance prediction of large-scale parallell system and application using macro-level simulation

Ryutaro Susukita, Yasunori Kimura, Hisashige Ando, Hidemi Komatsu, Mutsumi Aoyagi, Motoyoshi Kurokawa, Hiroaki Honda, Kazuaki J. Murakami, Yuichi Inadomi, Hidetomo Shibamura, Koji Inoue, Shuji Yamamura, Shigeru Ishizuki, Yunqing Yu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

23 Citations (Scopus)

Abstract

To predict application performance on an HPC system is an important technology for designing the computing system and developing applications. However, accurate prediction is a challenge, particularly, in the case of a future coming system with higher performance. In this paper, we present a new method for predicting application performance on HPC systems. This method combines modeling of sequential performance on a single processor and macro-level simulations of applications for parallel performance on the entire system. In the simulation, the execution flow is traced but kernel computations are omitted for reducing the execution time. Validation on a real terascale system showed that the predicted and measured performance agreed within 10% to 20 %. We employed the method in designing a hypothetical petascale system of 32768 SIMD-extended processor cores. For predicting application performance on the petascale system, the macro-level simulation required several hours.

Original languageEnglish
Title of host publication2008 SC - International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2008
DOIs
Publication statusPublished - Dec 1 2008
Event2008 SC - International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2008 - Austin, TX, United States
Duration: Nov 15 2008Nov 21 2008

Publication series

Name2008 SC - International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2008

Other

Other2008 SC - International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2008
CountryUnited States
CityAustin, TX
Period11/15/0811/21/08

Fingerprint

Macros
Large scale systems

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Software

Cite this

Susukita, R., Kimura, Y., Ando, H., Komatsu, H., Aoyagi, M., Kurokawa, M., ... Yu, Y. (2008). Performance prediction of large-scale parallell system and application using macro-level simulation. In 2008 SC - International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2008 [5220091] (2008 SC - International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2008). https://doi.org/10.1109/SC.2008.5220091

Performance prediction of large-scale parallell system and application using macro-level simulation. / Susukita, Ryutaro; Kimura, Yasunori; Ando, Hisashige; Komatsu, Hidemi; Aoyagi, Mutsumi; Kurokawa, Motoyoshi; Honda, Hiroaki; Murakami, Kazuaki J.; Inadomi, Yuichi; Shibamura, Hidetomo; Inoue, Koji; Yamamura, Shuji; Ishizuki, Shigeru; Yu, Yunqing.

2008 SC - International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2008. 2008. 5220091 (2008 SC - International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2008).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Susukita, R, Kimura, Y, Ando, H, Komatsu, H, Aoyagi, M, Kurokawa, M, Honda, H, Murakami, KJ, Inadomi, Y, Shibamura, H, Inoue, K, Yamamura, S, Ishizuki, S & Yu, Y 2008, Performance prediction of large-scale parallell system and application using macro-level simulation. in 2008 SC - International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2008., 5220091, 2008 SC - International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2008, 2008 SC - International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2008, Austin, TX, United States, 11/15/08. https://doi.org/10.1109/SC.2008.5220091
Susukita R, Kimura Y, Ando H, Komatsu H, Aoyagi M, Kurokawa M et al. Performance prediction of large-scale parallell system and application using macro-level simulation. In 2008 SC - International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2008. 2008. 5220091. (2008 SC - International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2008). https://doi.org/10.1109/SC.2008.5220091
Susukita, Ryutaro ; Kimura, Yasunori ; Ando, Hisashige ; Komatsu, Hidemi ; Aoyagi, Mutsumi ; Kurokawa, Motoyoshi ; Honda, Hiroaki ; Murakami, Kazuaki J. ; Inadomi, Yuichi ; Shibamura, Hidetomo ; Inoue, Koji ; Yamamura, Shuji ; Ishizuki, Shigeru ; Yu, Yunqing. / Performance prediction of large-scale parallell system and application using macro-level simulation. 2008 SC - International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2008. 2008. (2008 SC - International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2008).
@inproceedings{b100694183a44b8cb28b8f9d27105270,
title = "Performance prediction of large-scale parallell system and application using macro-level simulation",
abstract = "To predict application performance on an HPC system is an important technology for designing the computing system and developing applications. However, accurate prediction is a challenge, particularly, in the case of a future coming system with higher performance. In this paper, we present a new method for predicting application performance on HPC systems. This method combines modeling of sequential performance on a single processor and macro-level simulations of applications for parallel performance on the entire system. In the simulation, the execution flow is traced but kernel computations are omitted for reducing the execution time. Validation on a real terascale system showed that the predicted and measured performance agreed within 10{\%} to 20 {\%}. We employed the method in designing a hypothetical petascale system of 32768 SIMD-extended processor cores. For predicting application performance on the petascale system, the macro-level simulation required several hours.",
author = "Ryutaro Susukita and Yasunori Kimura and Hisashige Ando and Hidemi Komatsu and Mutsumi Aoyagi and Motoyoshi Kurokawa and Hiroaki Honda and Murakami, {Kazuaki J.} and Yuichi Inadomi and Hidetomo Shibamura and Koji Inoue and Shuji Yamamura and Shigeru Ishizuki and Yunqing Yu",
year = "2008",
month = "12",
day = "1",
doi = "10.1109/SC.2008.5220091",
language = "English",
isbn = "9781424428359",
series = "2008 SC - International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2008",
booktitle = "2008 SC - International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2008",

}

TY - GEN

T1 - Performance prediction of large-scale parallell system and application using macro-level simulation

AU - Susukita, Ryutaro

AU - Kimura, Yasunori

AU - Ando, Hisashige

AU - Komatsu, Hidemi

AU - Aoyagi, Mutsumi

AU - Kurokawa, Motoyoshi

AU - Honda, Hiroaki

AU - Murakami, Kazuaki J.

AU - Inadomi, Yuichi

AU - Shibamura, Hidetomo

AU - Inoue, Koji

AU - Yamamura, Shuji

AU - Ishizuki, Shigeru

AU - Yu, Yunqing

PY - 2008/12/1

Y1 - 2008/12/1

N2 - To predict application performance on an HPC system is an important technology for designing the computing system and developing applications. However, accurate prediction is a challenge, particularly, in the case of a future coming system with higher performance. In this paper, we present a new method for predicting application performance on HPC systems. This method combines modeling of sequential performance on a single processor and macro-level simulations of applications for parallel performance on the entire system. In the simulation, the execution flow is traced but kernel computations are omitted for reducing the execution time. Validation on a real terascale system showed that the predicted and measured performance agreed within 10% to 20 %. We employed the method in designing a hypothetical petascale system of 32768 SIMD-extended processor cores. For predicting application performance on the petascale system, the macro-level simulation required several hours.

AB - To predict application performance on an HPC system is an important technology for designing the computing system and developing applications. However, accurate prediction is a challenge, particularly, in the case of a future coming system with higher performance. In this paper, we present a new method for predicting application performance on HPC systems. This method combines modeling of sequential performance on a single processor and macro-level simulations of applications for parallel performance on the entire system. In the simulation, the execution flow is traced but kernel computations are omitted for reducing the execution time. Validation on a real terascale system showed that the predicted and measured performance agreed within 10% to 20 %. We employed the method in designing a hypothetical petascale system of 32768 SIMD-extended processor cores. For predicting application performance on the petascale system, the macro-level simulation required several hours.

UR - http://www.scopus.com/inward/record.url?scp=70350780320&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70350780320&partnerID=8YFLogxK

U2 - 10.1109/SC.2008.5220091

DO - 10.1109/SC.2008.5220091

M3 - Conference contribution

AN - SCOPUS:70350780320

SN - 9781424428359

T3 - 2008 SC - International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2008

BT - 2008 SC - International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2008

ER -