Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group
Conferences > OFC > 2023 > W1G > W1G.1

SiP Architecture For Accelerating Collective Communication in Distributed Deep Learning

Zhenguo Wu, Liang Yuan Dai, Ziyi Zhu, Asher Novick, Madeleine Glick, and Keren Bergman

Not Accessible

Your library or personal account may give you access

Abstract

We present a silicon photonic architecture for accelerating collective commu nications in distributed deep learning. We demonstrate a 22% job completion time improvement in a small-scale testbed and 1.4 to 5.9x improvement in large-scale simulations.

© 2023 The Author(s)

PDF Article  |   Presentation Video
More Like This
Integrating Nanosecond Optical Switching in Deep Distributed Learning System

Cen Wang, Noboru Yoshikane, and Takehiro Tsuritani
Th3D.5 Optical Fiber Communication Conference (OFC) 2023

Accelerating Distributed Machine Learning in Disaggregated Architectures with Flexible Optically Interconnected Computing Resources

Shijia Yan, Ziyi Zhu, Madeleine S. Glick, Zhenguo Wu, and Keren Bergman
Th1G.2 Optical Fiber Communication Conference (OFC) 2022

Which can Accelerate Distributed Machine Learning Faster: Hybrid Optical/Electrical or Optical Reconfigurable DCN?

Hao Yang, Zuqing Zhu, Roberto Proietti, and S. J. Ben Yoo
Th1G.5 Optical Fiber Communication Conference (OFC) 2022

View More...

Presentation Video

Presentation video access is available to:

  1. Optica Publishing Group subscribers
  2. Technical meeting attendees
  3. Optica members who wish to use one of their free downloads. Please download the article first. After downloading, please refresh this page.

Contact your librarian or system administrator
or
Log in to access Optica Member Subscription or free downloads

More Like This
Integrating Nanosecond Optical Switching in Deep Distributed Learning System

Cen Wang, Noboru Yoshikane, and Takehiro Tsuritani
Th3D.5 Optical Fiber Communication Conference (OFC) 2023

Accelerating Distributed Machine Learning in Disaggregated Architectures with Flexible Optically Interconnected Computing Resources

Shijia Yan, Ziyi Zhu, Madeleine S. Glick, Zhenguo Wu, and Keren Bergman
Th1G.2 Optical Fiber Communication Conference (OFC) 2022

Which can Accelerate Distributed Machine Learning Faster: Hybrid Optical/Electrical or Optical Reconfigurable DCN?

Hao Yang, Zuqing Zhu, Roberto Proietti, and S. J. Ben Yoo
Th1G.5 Optical Fiber Communication Conference (OFC) 2022

View More...
Silicon Photonic Switch-Enabled Server Regrouping Using Bandwidth Steering for Distributed Deep Learning Training

Ziyi Zhu, Shijia Yan, Madeleine Strom Glick, Min Yee Teh, and Keren Bergman
Th5H.3 Optical Fiber Communication Conference (OFC) 2021

Acceleration and Efficiency Warranty for Distributed Machine Learning Jobs over Data Center Network with Optical Circuit Switching

Cen Wang, Noboru Yoshikane, Filippos Balasis, and Takehiro Tsuritani
W1E.3 Optical Fiber Communication Conference (OFC) 2021

Select as filters
Select Topics Cancel
Top
       
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.
Privacy | Terms of Use