Tachyum Networking Strategy and Direction for Prodigy Data Center Deployments

  • Oct 26, 2022 Date of publishing
  • 25 Pages

Introduction

Tachyum is best known for its flagship product, Prodigy, the world’s first Universal Processor that unifies the functionality of CPU, GPGPU, and TPU into a single monolithic device.

The Prodigy Universal Processor delivers industry leading performance and efficiency for next generation supercomputers and requires cutting-edge networking infrastructure to complement the computing power for the supercomputer clusters. This paper begins with an overview of modern-day networking and how it has evolved since the invention of Ethernet and continues with the strategy that Tachyum has developed to provide the optimal networking infrastructure to support Prodigy compute nodes in upcoming supercomputer deployments.

The paper concludes with a glimpse into Tachyum’s future direction based on the trends in high performance networking technologies.

Figures

  1. Breakout of ToR Switch Ports
    Breakout of ToR Switch Ports
  2. 500PF Group Topology Diagram
    500PF Group Topology Diagram
  3. 1EF Group Topology Diagram with TH5 Switches
    1EF Group Topology Diagram with TH5 Switches
  4. 16EF Group Topology Diagram with TH5 Switches, Approach \#1
    16EF Group Topology Diagram with TH5 Switches, Approach \#1
  5. 16EF Group Topology Diagram with TH5 Switches, Approach \#2
    16EF Group Topology Diagram with TH5 Switches, Approach \#2
  6. 16EF Group Top-Level Diagram Illustrating Modular, Scalable Hierarchy
    16EF Group Top-Level Diagram Illustrating Modular, Scalable Hierarchy