Approaches for memory-efficient communication library and runtime communication optimization

Takeshi Nanri

研究成果: 著書/レポートタイプへの貢献 › 章

抄録

This article summarizes the works established in Advanced Communication for Exa (ACE) project. The most important motivation of this project was the severe demands for scalable communication toward Exa-scale computations. Therefore, in the project, we have built a PGAS-based communication library, Advanced Communication Primitives (ACP). Its fundamental communication model is onesided, based on PGAS model, so that it can consume internal memory footprint as small as possible. Based on this model, several applications including simulations of magnetohydrodynamic, molecular orbitals, and particles were tuned to achieve higher scalability. In addition to that, some communication optimization techniques have been investigated. Especially, tuning methods of collective communications, such as message ordering, algorithm selection, and overlapping, are studied. Also, in this project, a network simulator NSIM-ACE is developed. It simulates behavior of packets for one-sided communications to study the effects of congestions on interconnects.

元の言語英語
ホスト出版物のタイトルAdvanced Software Technologies for Post-Peta Scale Computing
ホスト出版物のサブタイトルThe Japanese Post-Peta CREST Research Project
出版者Springer Singapore
ページ121-138
ページ数18
ISBN（電子版）9789811319242
ISBN（印刷物）9789811319235
DOI
出版物ステータス出版済み - 12 6 2018

Fingerprint

Data storage equipment
Communication
Molecular orbitals
Magnetohydrodynamics
Scalability
Tuning
Simulators

All Science Journal Classification (ASJC) codes

  • Computer Science(all)

これを引用

Nanri, T. (2018). Approaches for memory-efficient communication library and runtime communication optimization. ： Advanced Software Technologies for Post-Peta Scale Computing: The Japanese Post-Peta CREST Research Project (pp. 121-138). Springer Singapore. https://doi.org/10.1007/978-981-13-1924-2_7

Approaches for memory-efficient communication library and runtime communication optimization. / Nanri, Takeshi.

Advanced Software Technologies for Post-Peta Scale Computing: The Japanese Post-Peta CREST Research Project. Springer Singapore, 2018. p. 121-138.

研究成果: 著書/レポートタイプへの貢献 › 章

Nanri, T 2018, Approaches for memory-efficient communication library and runtime communication optimization. ： Advanced Software Technologies for Post-Peta Scale Computing: The Japanese Post-Peta CREST Research Project. Springer Singapore, pp. 121-138. https://doi.org/10.1007/978-981-13-1924-2_7
Nanri T. Approaches for memory-efficient communication library and runtime communication optimization. ： Advanced Software Technologies for Post-Peta Scale Computing: The Japanese Post-Peta CREST Research Project. Springer Singapore. 2018. p. 121-138 https://doi.org/10.1007/978-981-13-1924-2_7
Nanri, Takeshi. / Approaches for memory-efficient communication library and runtime communication optimization. Advanced Software Technologies for Post-Peta Scale Computing: The Japanese Post-Peta CREST Research Project. Springer Singapore, 2018. pp. 121-138
@inbook{95d0663e97664296bc0c13c8f116f67c,
title = "Approaches for memory-efficient communication library and runtime communication optimization",
abstract = "This article summarizes the works established in Advanced Communication for Exa (ACE) project. The most important motivation of this project was the severe demands for scalable communication toward Exa-scale computations. Therefore, in the project, we have built a PGAS-based communication library, Advanced Communication Primitives (ACP). Its fundamental communication model is onesided, based on PGAS model, so that it can consume internal memory footprint as small as possible. Based on this model, several applications including simulations of magnetohydrodynamic, molecular orbitals, and particles were tuned to achieve higher scalability. In addition to that, some communication optimization techniques have been investigated. Especially, tuning methods of collective communications, such as message ordering, algorithm selection, and overlapping, are studied. Also, in this project, a network simulator NSIM-ACE is developed. It simulates behavior of packets for one-sided communications to study the effects of congestions on interconnects.",
author = "Takeshi Nanri",
year = "2018",
month = "12",
day = "6",
doi = "10.1007/978-981-13-1924-2_7",
language = "English",
isbn = "9789811319235",
pages = "121--138",
booktitle = "Advanced Software Technologies for Post-Peta Scale Computing",
publisher = "Springer Singapore",

}

TY - CHAP

T1 - Approaches for memory-efficient communication library and runtime communication optimization

AU - Nanri, Takeshi

PY - 2018/12/6

Y1 - 2018/12/6

N2 - This article summarizes the works established in Advanced Communication for Exa (ACE) project. The most important motivation of this project was the severe demands for scalable communication toward Exa-scale computations. Therefore, in the project, we have built a PGAS-based communication library, Advanced Communication Primitives (ACP). Its fundamental communication model is onesided, based on PGAS model, so that it can consume internal memory footprint as small as possible. Based on this model, several applications including simulations of magnetohydrodynamic, molecular orbitals, and particles were tuned to achieve higher scalability. In addition to that, some communication optimization techniques have been investigated. Especially, tuning methods of collective communications, such as message ordering, algorithm selection, and overlapping, are studied. Also, in this project, a network simulator NSIM-ACE is developed. It simulates behavior of packets for one-sided communications to study the effects of congestions on interconnects.

AB - This article summarizes the works established in Advanced Communication for Exa (ACE) project. The most important motivation of this project was the severe demands for scalable communication toward Exa-scale computations. Therefore, in the project, we have built a PGAS-based communication library, Advanced Communication Primitives (ACP). Its fundamental communication model is onesided, based on PGAS model, so that it can consume internal memory footprint as small as possible. Based on this model, several applications including simulations of magnetohydrodynamic, molecular orbitals, and particles were tuned to achieve higher scalability. In addition to that, some communication optimization techniques have been investigated. Especially, tuning methods of collective communications, such as message ordering, algorithm selection, and overlapping, are studied. Also, in this project, a network simulator NSIM-ACE is developed. It simulates behavior of packets for one-sided communications to study the effects of congestions on interconnects.

UR - http://www.scopus.com/inward/record.url?scp=85063805105&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85063805105&partnerID=8YFLogxK

U2 - 10.1007/978-981-13-1924-2_7

DO - 10.1007/978-981-13-1924-2_7

M3 - Chapter

AN - SCOPUS:85063805105

SN - 9789811319235

SP - 121

EP - 138

BT - Advanced Software Technologies for Post-Peta Scale Computing

PB - Springer Singapore

ER -

文献にアクセス