Machine Learning aided Computer Architecture Design for CNN Inferencing Systems

TL;DR

This paper introduces a rapid, accurate method for predicting power and performance of CNN inference on GPGPUs, streamlining the design space exploration process for efficient hardware selection in ML systems.

Contribution

It presents a novel technique for fast power and performance forecasting of CNNs on GPGPUs, reducing manual effort and development time.

Findings

MAPE of 5.03% for power prediction

MAPE of 5.94% for performance prediction

Enables early-stage estimation to save time and costs

Abstract

Efficient and timely calculations of Machine Learning (ML) algorithms are essential for emerging technologies like autonomous driving, the Internet of Things (IoT), and edge computing. One of the primary ML algorithms used in such systems is Convolutional Neural Networks (CNNs), which demand high computational resources. This requirement has led to the use of ML accelerators like GPGPUs to meet design constraints. However, selecting the most suitable accelerator involves Design Space Exploration (DSE), a process that is usually time-consuming and requires significant manual effort. Our work presents approaches to expedite the DSE process by identifying the most appropriate GPGPU for CNN inferencing systems. We have developed a quick and precise technique for forecasting the power and performance of CNNs during inference, with a MAPE of 5.03% and 5.94%, respectively. Our approach empowers computer architects to estimate power and performance in the early stages of development, reducing the necessity for numerous prototypes. This saves time and money while also improving the time-to-market period.