Survey and Benchmarking of Machine Learning Accelerators

TL;DR

This paper surveys the landscape of machine learning accelerators, analyzes trends in performance and power, and benchmarks two low-power accelerators for embedded inference applications.

Contribution

It provides a comprehensive survey of publicly announced ML accelerators, visualizes performance trends, and benchmarks two promising low-power accelerators for embedded use.

Findings

Power consumption and numerical precision trends analyzed

Benchmark results show real-world performance of low-power accelerators

Comparison with Intel CPU highlights strengths and weaknesses

Abstract

Advances in multicore processors and accelerators have opened the flood gates to greater exploration and application of machine learning techniques to a variety of applications. These advances, along with breakdowns of several trends including Moore's Law, have prompted an explosion of processors and accelerators that promise even greater computational and machine learning capabilities. These processors and accelerators are coming in many forms, from CPUs and GPUs to ASICs, FPGAs, and dataflow accelerators. This paper surveys the current state of these processors and accelerators that have been publicly announced with performance and power consumption numbers. The performance and power values are plotted on a scatter graph and a number of dimensions and observations from the trends on this plot are discussed and analyzed. For instance, there are interesting trends in the plot regarding power consumption, numerical precision, and inference versus training. We then select and benchmark two commercially-available low size, weight, and power (SWaP) accelerators as these processors are the most interesting for embedded and mobile machine learning inference applications that are most applicable to the DoD and other SWaP constrained users. We determine how they actually perform with real-world images and neural network models, compare those results to the reported performance and power consumption values and evaluate them against an Intel CPU that is used in some embedded applications.