Efficient intranode communication in GPU-accelerated systems

Feng Ji; Ashwin M. Aji; James Dinan; Darius Buntinas; Pavan Balaji; Wu Chun Feng; Xiaosong Ma

doi:10.1109/IPDPSW.2012.227

Efficient intranode communication in GPU-accelerated systems

Feng Ji^*, Ashwin M. Aji, James Dinan, Darius Buntinas, Pavan Balaji, Wu Chun Feng, Xiaosong Ma

^*Corresponding author for this work

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

12 Citations (Scopus)

Abstract

Current implementations of MPI are unaware of accelerator memory (i.e., GPU device memory) and require programmers to explicitly move data between memory spaces. This approach is inefficient, especially for intranode communication where it can result in several extra copy operations. In this work, we integrate GPU-awareness into a popular MPI runtime system and develop techniques to significantly reduce the cost of intranode communication involving one or more GPUs. Experiment results show an up to 2x increase in bandwidth, resulting in an average of 4.3% improvement to the total execution time of a halo exchange benchmark.

Original language	English
Title of host publication	Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012
Pages	1838-1847
Number of pages	10
DOIs	https://doi.org/10.1109/IPDPSW.2012.227
Publication status	Published - 2012
Externally published	Yes
Event	2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012 - Shanghai, China Duration: 21 May 2012 → 25 May 2012

Publication series

Name	Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012

Conference

Conference	2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012
Country/Territory	China
City	Shanghai
Period	21/05/12 → 25/05/12

Keywords

CUDA
GPU
Intranode communication
MPI
MPICH2
Nemesis

Access to Document

10.1109/IPDPSW.2012.227

Cite this

Ji, F., Aji, A. M., Dinan, J., Buntinas, D., Balaji, P., Feng, W. C., & Ma, X. (2012). Efficient intranode communication in GPU-accelerated systems. In Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012 (pp. 1838-1847). Article 6270862 (Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012). https://doi.org/10.1109/IPDPSW.2012.227

@inproceedings{6234a729196b44ee82c213aa6cd36410,

title = "Efficient intranode communication in GPU-accelerated systems",

abstract = "Current implementations of MPI are unaware of accelerator memory (i.e., GPU device memory) and require programmers to explicitly move data between memory spaces. This approach is inefficient, especially for intranode communication where it can result in several extra copy operations. In this work, we integrate GPU-awareness into a popular MPI runtime system and develop techniques to significantly reduce the cost of intranode communication involving one or more GPUs. Experiment results show an up to 2x increase in bandwidth, resulting in an average of 4.3% improvement to the total execution time of a halo exchange benchmark.",

keywords = "CUDA, GPU, Intranode communication, MPI, MPICH2, Nemesis",

author = "Feng Ji and Aji, {Ashwin M.} and James Dinan and Darius Buntinas and Pavan Balaji and Feng, {Wu Chun} and Xiaosong Ma",

year = "2012",

doi = "10.1109/IPDPSW.2012.227",

language = "English",

isbn = "9780769546766",

series = "Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012",

pages = "1838--1847",

booktitle = "Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012",

note = "2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012 ; Conference date: 21-05-2012 Through 25-05-2012",

}

Ji, F, Aji, AM, Dinan, J, Buntinas, D, Balaji, P, Feng, WC & Ma, X 2012, Efficient intranode communication in GPU-accelerated systems. in Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012., 6270862, Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012, pp. 1838-1847, 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012, Shanghai, China, 21/05/12. https://doi.org/10.1109/IPDPSW.2012.227

Efficient intranode communication in GPU-accelerated systems. / Ji, Feng; Aji, Ashwin M.; Dinan, James et al.
Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012. 2012. p. 1838-1847 6270862 (Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012).

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

TY - GEN

T1 - Efficient intranode communication in GPU-accelerated systems

AU - Ji, Feng

AU - Aji, Ashwin M.

AU - Dinan, James

AU - Buntinas, Darius

AU - Balaji, Pavan

AU - Feng, Wu Chun

AU - Ma, Xiaosong

PY - 2012

Y1 - 2012

N2 - Current implementations of MPI are unaware of accelerator memory (i.e., GPU device memory) and require programmers to explicitly move data between memory spaces. This approach is inefficient, especially for intranode communication where it can result in several extra copy operations. In this work, we integrate GPU-awareness into a popular MPI runtime system and develop techniques to significantly reduce the cost of intranode communication involving one or more GPUs. Experiment results show an up to 2x increase in bandwidth, resulting in an average of 4.3% improvement to the total execution time of a halo exchange benchmark.

AB - Current implementations of MPI are unaware of accelerator memory (i.e., GPU device memory) and require programmers to explicitly move data between memory spaces. This approach is inefficient, especially for intranode communication where it can result in several extra copy operations. In this work, we integrate GPU-awareness into a popular MPI runtime system and develop techniques to significantly reduce the cost of intranode communication involving one or more GPUs. Experiment results show an up to 2x increase in bandwidth, resulting in an average of 4.3% improvement to the total execution time of a halo exchange benchmark.

KW - CUDA

KW - GPU

KW - Intranode communication

KW - MPI

KW - MPICH2

KW - Nemesis

UR - http://www.scopus.com/inward/record.url?scp=84867431904&partnerID=8YFLogxK

U2 - 10.1109/IPDPSW.2012.227

DO - 10.1109/IPDPSW.2012.227

M3 - Conference contribution

AN - SCOPUS:84867431904

SN - 9780769546766

T3 - Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012

SP - 1838

EP - 1847

BT - Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012

T2 - 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012

Y2 - 21 May 2012 through 25 May 2012

ER -

Ji F, Aji AM, Dinan J, Buntinas D, Balaji P, Feng WC et al. Efficient intranode communication in GPU-accelerated systems. In Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012. 2012. p. 1838-1847. 6270862. (Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012). doi: 10.1109/IPDPSW.2012.227

Efficient intranode communication in GPU-accelerated systems

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this