Link shutdown opportunities during collective communications in 3-D torus nets

S. Conner; S. Akioka; M. J. Irwin; P. Raghavan

doi:10.1109/IPDPS.2007.370534

Link shutdown opportunities during collective communications in 3-D torus nets

S. Conner
, S. Akioka
, M. J. Irwin
, P. Raghavan

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

8 Scopus citations

Abstract

As modern computing clusters used in scientific computing applications scale to ever-larger sizes and capabilities, their operational energy costs have become prohibitive. While it is an emerging trend in modern cluster design to optimize for low energy consumption in the individual computational nodes, little attention has been paid to reducing the energy used by the communication network that connects the nodes. In this work, we consider a 3-D torus network similar to the one in BlueGene/L to explore opporunities for link shutdown during collective communication operations. For example, we demonstrate that in the case of all-to-one reduce codes, approximately 99% of the total network link time can be spent in a shutoff state on a 64-node toroidal network, thus reducing the overall system energy by approximately 15-28%

Original language	English (US)
Title of host publication	Proceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM
DOIs	https://doi.org/10.1109/IPDPS.2007.370534
State	Published - 2007
Event	21st International Parallel and Distributed Processing Symposium, IPDPS 2007 - Long Beach, CA, United States Duration: Mar 26 2007 → Mar 30 2007

Publication series

Name	Proceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM

Other

Other	21st International Parallel and Distributed Processing Symposium, IPDPS 2007
Country/Territory	United States
City	Long Beach, CA
Period	3/26/07 → 3/30/07

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 7 Affordable and Clean Energy

All Science Journal Classification (ASJC) codes

Hardware and Architecture
Software
General Mathematics

Access to Document

10.1109/IPDPS.2007.370534

Cite this

Conner, S., Akioka, S., Irwin, M. J., & Raghavan, P. (2007). Link shutdown opportunities during collective communications in 3-D torus nets. In Proceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM Article 4228262 (Proceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM). https://doi.org/10.1109/IPDPS.2007.370534

@inproceedings{bf34e3c0476e451682e6710ff35ff2e2,

title = "Link shutdown opportunities during collective communications in 3-D torus nets",

abstract = "As modern computing clusters used in scientific computing applications scale to ever-larger sizes and capabilities, their operational energy costs have become prohibitive. While it is an emerging trend in modern cluster design to optimize for low energy consumption in the individual computational nodes, little attention has been paid to reducing the energy used by the communication network that connects the nodes. In this work, we consider a 3-D torus network similar to the one in BlueGene/L to explore opporunities for link shutdown during collective communication operations. For example, we demonstrate that in the case of all-to-one reduce codes, approximately 99\% of the total network link time can be spent in a shutoff state on a 64-node toroidal network, thus reducing the overall system energy by approximately 15-28\%",

author = "S. Conner and S. Akioka and Irwin, \{M. J.\} and P. Raghavan",

year = "2007",

doi = "10.1109/IPDPS.2007.370534",

language = "English (US)",

isbn = "1424409101",

series = "Proceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM",

booktitle = "Proceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM",

note = "21st International Parallel and Distributed Processing Symposium, IPDPS 2007 ; Conference date: 26-03-2007 Through 30-03-2007",

}

Conner, S, Akioka, S, Irwin, MJ & Raghavan, P 2007, Link shutdown opportunities during collective communications in 3-D torus nets. in Proceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM., 4228262, Proceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM, 21st International Parallel and Distributed Processing Symposium, IPDPS 2007, Long Beach, CA, United States, 3/26/07. https://doi.org/10.1109/IPDPS.2007.370534

Link shutdown opportunities during collective communications in 3-D torus nets. / Conner, S.; Akioka, S.; Irwin, M. J. et al.
Proceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM. 2007. 4228262 (Proceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM).

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

TY - GEN

T1 - Link shutdown opportunities during collective communications in 3-D torus nets

AU - Conner, S.

AU - Akioka, S.

AU - Irwin, M. J.

AU - Raghavan, P.

PY - 2007

Y1 - 2007

N2 - As modern computing clusters used in scientific computing applications scale to ever-larger sizes and capabilities, their operational energy costs have become prohibitive. While it is an emerging trend in modern cluster design to optimize for low energy consumption in the individual computational nodes, little attention has been paid to reducing the energy used by the communication network that connects the nodes. In this work, we consider a 3-D torus network similar to the one in BlueGene/L to explore opporunities for link shutdown during collective communication operations. For example, we demonstrate that in the case of all-to-one reduce codes, approximately 99% of the total network link time can be spent in a shutoff state on a 64-node toroidal network, thus reducing the overall system energy by approximately 15-28%

AB - As modern computing clusters used in scientific computing applications scale to ever-larger sizes and capabilities, their operational energy costs have become prohibitive. While it is an emerging trend in modern cluster design to optimize for low energy consumption in the individual computational nodes, little attention has been paid to reducing the energy used by the communication network that connects the nodes. In this work, we consider a 3-D torus network similar to the one in BlueGene/L to explore opporunities for link shutdown during collective communication operations. For example, we demonstrate that in the case of all-to-one reduce codes, approximately 99% of the total network link time can be spent in a shutoff state on a 64-node toroidal network, thus reducing the overall system energy by approximately 15-28%

UR - https://www.scopus.com/pages/publications/34548738126

UR - https://www.scopus.com/pages/publications/34548738126#tab=citedBy

U2 - 10.1109/IPDPS.2007.370534

DO - 10.1109/IPDPS.2007.370534

M3 - Conference contribution

AN - SCOPUS:34548738126

SN - 1424409101

SN - 9781424409105

T3 - Proceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM

BT - Proceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM

T2 - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007

Y2 - 26 March 2007 through 30 March 2007

ER -

Conner S, Akioka S, Irwin MJ, Raghavan P. Link shutdown opportunities during collective communications in 3-D torus nets. In Proceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM. 2007. 4228262. (Proceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM). doi: 10.1109/IPDPS.2007.370534

Link shutdown opportunities during collective communications in 3-D torus nets

Abstract

Publication series

Other

UN SDGs

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this