The Data Conversion Bottleneck in Analog Computing Accelerators

TL;DR

This paper investigates the limitations of analog optical accelerators caused by data conversion bottlenecks, and presents a case study showing potential speedups for specific applications like Fourier transforms and convolutions.

Contribution

It provides a detailed case study of an optical Fourier transform and convolution accelerator, highlighting the performance benefits and limitations due to data conversion overheads.

Findings

Average speedup of 9.4 times across benchmarks

Significant speedup for pure Fourier transform and convolution tasks

Data conversion costs limit overall performance gains

Abstract

Most modern computing tasks have digital electronic input and output data. Due to these constraints imposed by real-world use cases of computer systems, any analog computing accelerator, whether analog electronic or optical, must perform an analog-to-digital conversion on its input data and a subsequent digital-to-analog conversion on its output data. The energy and latency costs incurred by data conversion place performance limits on analog computing accelerators. To avoid this overhead, analog hardware must replace the full functionality of traditional digital electronic computer hardware. This is not currently possible for optical computing accelerators due to limitations in gain, input-output isolation, and information storage in optical hardware. This article presents a case study that profiles 27 benchmarks for an analog optical Fourier transform and convolution accelerator which we designed and built. The case study shows that an ideal optical Fourier transform and convolution accelerator can produce an average speedup of 9.4 times and a median speedup of 1.9 times for the set of benchmarks. The optical Fourier transform and convolution accelerator only produces significant speedup for pure Fourier transform (45.3 times) and convolution (159.4 times) applications.