Back to Search Start Over

An Electro-Photonic System for Accelerating Deep Neural Networks.

Authors :
DEMIRKIRAN, CANSU
ERIS, FURKAN
GONGYU WANG
ELMHURST, JONATHAN
MOORE, NICK
HARRIS, NICHOLAS C.
BASUMALLIK, AYON
REDDI, VIJAY JANAPA
JOSHI, AJAY
BUNANDAR, DARIUS
Source :
ACM Journal on Emerging Technologies in Computing Systems; Oct2023, Vol. 19 Issue 4, p1-31, 31p
Publication Year :
2023

Abstract

The number of parameters in deep neural networks (DNNs) is scaling at about 5× the rate of Moore's Law. To sustain this growth, photonic computing is a promising avenue, as it enables higher throughput in dominant general matrix-matrix multiplication (GEMM) operations in DNNs than their electrical counterpart. However, purely photonic systems face several challenges including lack of photonic memory and accumulation of noise. In this article, we present an electro-photonic accelerator, ADEPT, which leverages a photonic computing unit for performing GEMM operations, a vectorized digital electronic application-specific integrated circuits for performing non-GEMM operations, and SRAM arrays for storing DNN parameters and activations. In contrast to prior works in photonic DNN accelerators, we adopt a system-level perspective and show that the gains while large are tempered relative to prior expectations. Our goal is to encourage architects to explore photonic technology in a more pragmatic way considering the system as a whole to understand its general applicability in accelerating today's DNNs. Our evaluation shows that ADEPT can provide, on average, 5.73× higher throughput per watt compared to the traditional systolic arrays in a full-system, and at least 6.8× and 2.5× better throughput per watt, compared to state-of-the-art electronic and photonic accelerators, respectively. [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
15504832
Volume :
19
Issue :
4
Database :
Complementary Index
Journal :
ACM Journal on Emerging Technologies in Computing Systems
Publication Type :
Academic Journal
Accession number :
173770618
Full Text :
https://doi.org/10.1145/3606949