Photonics in the Package for Extreme Scalability (PIPES)

Abstract

The Photonics in the Package for Extreme Scalability (PIPES) program aims to develop optical signaling technologies for digital microelectronics. Distributed and parallel computing architectures are now pervasive across all size scales, from personal-scale multicore processing units to enterprise-scale high performance computing systems, and span application domains from consumer electronics to DoD systems. Increasingly, however, the benefits of parallelism are constrained not by the limits of computation at individual nodes but by the movement of data between nodes. PIPES will advance microelectronics capabilities by intimately integrating photonics with advanced integrated electronics to yield system connectivity with an unprecedented combination of high aggregate bandwidth, power efficiency, channel density, and link reach. Specifically, PIPES will develop photonic input/output (I/O) capability for application-specific integrated circuits and Field-Programmable Gate Arrays (FPGAs) that are widely used in advanced DoD sensors and radio frequency systems. The goal of the program is improving I/O bandwidth density, efficiency, and reach by more than a factor of 100 to enable disruptive DoD system parallelism and performance scaling. As PIPES technologies mature, they are anticipated to proliferate into central processing units, graphical processing units, and emerging tensor-flow processing units that will impact a wide range of dual-use applications including artificial intelligence, machine learning, large scale emulation, and high performance computing. Technologies from this program are intended for transition to larger scale commercial performers and the Services.

Document Details

Document Type
Accomplishment
Publication Date
Oct 01, 2022
Source ID
3cb6489e12b19dd39f6dc7dd8d43a1de

Tags

Fields of Study

  • Engineering

Readers

  • Distributed Systems and Data Platform Development
  • Integrated Circuit Design and Technology.
  • Parallel and Distributed Computing.

Technology Areas

  • AI & ML
  • AI & ML - DoD AI Strategy
  • Microelectronics

Related Documents