RSON: An inter/intra-chip silicon photonic network for rack-scale computing systems

Peng Yang; Zhengbin Pang; Zhifei Wang; Zhehui Wang; Min Xie; Xuanqi Chen; Luan H.K. Duong; Jiang Xu

doi:10.23919/DATE.2018.8342226

RSON: An inter/intra-chip silicon photonic network for rack-scale computing systems

Peng Yang
, Zhengbin Pang
, Zhifei Wang
, Zhehui Wang
, Min Xie
, Xuanqi Chen
, Luan H.K. Duong
, Jiang Xu

Department of Electronic and Computer Engineering

Research output: Chapter in Book/Conference Proceeding/Report › Conference Paper published in a book › peer-review

7 Citations (Scopus)

Abstract

The increasing demand for more computational power from scientific computing, big data processing, and machine learning is pushing the development of HPC (high-performance computing) systems. As the basic HPC building blocks, modularized server racks with a large number of multicore nodes are facing performance and energy efficiency challenges. This paper proposes RSON, an optical network for rack-scale computing systems. RSON connects processor cores, caches, local memories, and remote memories through a novel inter/intra-chip silicon photonic network architecture. We develop a low-latency scalable channel partition and low-power dynamic path priority control scheme for RSON. Experimental results show that RSON can help rack-scale computing systems achieve up to 6.8X higher performance under the same energy consumption than state-of-the-art systems under the latest APEX (application performance at extreme scale) benchmarks.

Original language	English
Title of host publication	Proceedings of the 2018 Design, Automation and Test in Europe Conference and Exhibition, DATE 2018
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1369-1374
Number of pages	6
ISBN (Electronic)	9783981926316
DOIs	https://doi.org/10.23919/DATE.2018.8342226
Publication status	Published - 19 Apr 2018
Event	2018 Design, Automation and Test in Europe Conference and Exhibition, DATE 2018 - Dresden, Germany Duration: 19 Mar 2018 → 23 Mar 2018

Publication series

Name	Proceedings of the 2018 Design, Automation and Test in Europe Conference and Exhibition, DATE 2018
Volume	2018-January

Conference

Conference	2018 Design, Automation and Test in Europe Conference and Exhibition, DATE 2018
Country/Territory	Germany
City	Dresden
Period	19/03/18 → 23/03/18

Bibliographical note

Publisher Copyright:
© 2018 EDAA.

UN SDGs

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

SDG 7 Affordable and Clean Energy

Access to Document

10.23919/DATE.2018.8342226

Cite this

Yang, P., Pang, Z., Wang, Z., Wang, Z., Xie, M., Chen, X., Duong, L. H. K., & Xu, J. (2018). RSON: An inter/intra-chip silicon photonic network for rack-scale computing systems. In Proceedings of the 2018 Design, Automation and Test in Europe Conference and Exhibition, DATE 2018 (pp. 1369-1374). (Proceedings of the 2018 Design, Automation and Test in Europe Conference and Exhibition, DATE 2018; Vol. 2018-January). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.23919/DATE.2018.8342226

Yang, Peng ; Pang, Zhengbin ; Wang, Zhifei et al. / RSON : An inter/intra-chip silicon photonic network for rack-scale computing systems. Proceedings of the 2018 Design, Automation and Test in Europe Conference and Exhibition, DATE 2018. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 1369-1374 (Proceedings of the 2018 Design, Automation and Test in Europe Conference and Exhibition, DATE 2018).

@inproceedings{5ad253a7f7ba44cd8f7e27167d8a3848,

title = "RSON: An inter/intra-chip silicon photonic network for rack-scale computing systems",

abstract = "The increasing demand for more computational power from scientific computing, big data processing, and machine learning is pushing the development of HPC (high-performance computing) systems. As the basic HPC building blocks, modularized server racks with a large number of multicore nodes are facing performance and energy efficiency challenges. This paper proposes RSON, an optical network for rack-scale computing systems. RSON connects processor cores, caches, local memories, and remote memories through a novel inter/intra-chip silicon photonic network architecture. We develop a low-latency scalable channel partition and low-power dynamic path priority control scheme for RSON. Experimental results show that RSON can help rack-scale computing systems achieve up to 6.8X higher performance under the same energy consumption than state-of-the-art systems under the latest APEX (application performance at extreme scale) benchmarks.",

author = "Peng Yang and Zhengbin Pang and Zhifei Wang and Zhehui Wang and Min Xie and Xuanqi Chen and Duong, \{Luan H.K.\} and Jiang Xu",

note = "Publisher Copyright: {\textcopyright} 2018 EDAA.; 2018 Design, Automation and Test in Europe Conference and Exhibition, DATE 2018 ; Conference date: 19-03-2018 Through 23-03-2018",

year = "2018",

month = apr,

day = "19",

doi = "10.23919/DATE.2018.8342226",

language = "English",

series = "Proceedings of the 2018 Design, Automation and Test in Europe Conference and Exhibition, DATE 2018",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "1369--1374",

booktitle = "Proceedings of the 2018 Design, Automation and Test in Europe Conference and Exhibition, DATE 2018",

}

Yang, P, Pang, Z, Wang, Z, Wang, Z, Xie, M, Chen, X, Duong, LHK & Xu, J 2018, RSON: An inter/intra-chip silicon photonic network for rack-scale computing systems. in Proceedings of the 2018 Design, Automation and Test in Europe Conference and Exhibition, DATE 2018. Proceedings of the 2018 Design, Automation and Test in Europe Conference and Exhibition, DATE 2018, vol. 2018-January, Institute of Electrical and Electronics Engineers Inc., pp. 1369-1374, 2018 Design, Automation and Test in Europe Conference and Exhibition, DATE 2018, Dresden, Germany, 19/03/18. https://doi.org/10.23919/DATE.2018.8342226

RSON: An inter/intra-chip silicon photonic network for rack-scale computing systems. / Yang, Peng; Pang, Zhengbin; Wang, Zhifei et al.
Proceedings of the 2018 Design, Automation and Test in Europe Conference and Exhibition, DATE 2018. Institute of Electrical and Electronics Engineers Inc., 2018. p. 1369-1374 (Proceedings of the 2018 Design, Automation and Test in Europe Conference and Exhibition, DATE 2018; Vol. 2018-January).

Research output: Chapter in Book/Conference Proceeding/Report › Conference Paper published in a book › peer-review

TY - GEN

T1 - RSON

T2 - 2018 Design, Automation and Test in Europe Conference and Exhibition, DATE 2018

AU - Yang, Peng

AU - Pang, Zhengbin

AU - Wang, Zhifei

AU - Wang, Zhehui

AU - Xie, Min

AU - Chen, Xuanqi

AU - Duong, Luan H.K.

AU - Xu, Jiang

PY - 2018/4/19

Y1 - 2018/4/19

N2 - The increasing demand for more computational power from scientific computing, big data processing, and machine learning is pushing the development of HPC (high-performance computing) systems. As the basic HPC building blocks, modularized server racks with a large number of multicore nodes are facing performance and energy efficiency challenges. This paper proposes RSON, an optical network for rack-scale computing systems. RSON connects processor cores, caches, local memories, and remote memories through a novel inter/intra-chip silicon photonic network architecture. We develop a low-latency scalable channel partition and low-power dynamic path priority control scheme for RSON. Experimental results show that RSON can help rack-scale computing systems achieve up to 6.8X higher performance under the same energy consumption than state-of-the-art systems under the latest APEX (application performance at extreme scale) benchmarks.

AB - The increasing demand for more computational power from scientific computing, big data processing, and machine learning is pushing the development of HPC (high-performance computing) systems. As the basic HPC building blocks, modularized server racks with a large number of multicore nodes are facing performance and energy efficiency challenges. This paper proposes RSON, an optical network for rack-scale computing systems. RSON connects processor cores, caches, local memories, and remote memories through a novel inter/intra-chip silicon photonic network architecture. We develop a low-latency scalable channel partition and low-power dynamic path priority control scheme for RSON. Experimental results show that RSON can help rack-scale computing systems achieve up to 6.8X higher performance under the same energy consumption than state-of-the-art systems under the latest APEX (application performance at extreme scale) benchmarks.

UR - https://www.scopus.com/pages/publications/85048789734

U2 - 10.23919/DATE.2018.8342226

DO - 10.23919/DATE.2018.8342226

M3 - Conference Paper published in a book

AN - SCOPUS:85048789734

T3 - Proceedings of the 2018 Design, Automation and Test in Europe Conference and Exhibition, DATE 2018

SP - 1369

EP - 1374

BT - Proceedings of the 2018 Design, Automation and Test in Europe Conference and Exhibition, DATE 2018

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 19 March 2018 through 23 March 2018

ER -

Yang P, Pang Z, Wang Z, Wang Z, Xie M, Chen X et al. RSON: An inter/intra-chip silicon photonic network for rack-scale computing systems. In Proceedings of the 2018 Design, Automation and Test in Europe Conference and Exhibition, DATE 2018. Institute of Electrical and Electronics Engineers Inc. 2018. p. 1369-1374. (Proceedings of the 2018 Design, Automation and Test in Europe Conference and Exhibition, DATE 2018). doi: 10.23919/DATE.2018.8342226

RSON: An inter/intra-chip silicon photonic network for rack-scale computing systems

Abstract

Publication series

Conference

Bibliographical note

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this