Distributed deep learning training using silicon photonic switched architectures

Abstract

The scaling trends of deep learning models and distributed training workloads are challenging network capacities in today’s datacenters and high-performance computing (HPC) systems. We propose a system architecture that leverages silicon photonic (SiP) switch-enabled server regrouping using bandwidth steering to tackle the challenges and accelerate distributed deep learning training. In addition, our proposed system architecture utilizes a highly integrated operating system-based SiP switch control scheme to reduce implementation complexity. To demonstrate the feasibility of our proposal, we built an experimental testbed with a SiP switch-enabled reconfigurable fat tree topology and evaluated the network performance of distributed ring all-reduce and parameter server workloads. The experimental results show up to 3.6× improvements over the static non-reconfigurable fat tree. Our large-scale simulation results show that server regrouping can deliver up to 2.3× flow throughput improvement for a 2× tapered fat tree and a further 11% improvement when higher-layer bandwidth steering is employed. The collective results show the potential of integrating SiP switches into datacenters and HPC systems to accelerate distributed deep learning training.

Document Details

Document Type
Pub Defense Publication
Publication Date
Mar 01, 2022
Source ID
10.1063/5.0070711

Entities

People

  • Keren Bergman
  • Maarten Hattink
  • Madeleine Strom Glick
  • Min Yee Teh
  • Shijia Yan
  • Zhenguo Wu
  • Ziyi Zhu

Organizations

  • Columbia University

Tags

Fields of Study

  • Computer science

Readers

  • Computer Networking
  • Phased Array Antenna Design.
  • STEM Education

Technology Areas

  • AI & ML
  • AI & ML - Neural Networks