Neural Network Inference on Mobile SoCs

TL;DR

This paper evaluates the inference capabilities and power-performance trade-offs of various components in mobile SoCs, demonstrating that parallel engagement of all components can double inference performance.

Contribution

It provides a comprehensive quantitative analysis of different ML components in mobile SoCs and explores the potential of concurrent multi-component inference for performance gains.

Findings

Up to 2x inference speedup with all components engaged

Different components exhibit distinct power-performance profiles

Parallel inference leverages heterogeneous SoC resources effectively

Abstract

The ever-increasing demand from mobile Machine Learning (ML) applications calls for evermore powerful on-chip computing resources. Mobile devices are empowered with heterogeneous multi-processor Systems-on-Chips (SoCs) to process ML workloads such as Convolutional Neural Network (CNN) inference. Mobile SoCs house several different types of ML capable components on-die, such as CPU, GPU, and accelerators. These different components are capable of independently performing inference but with very different power-performance characteristics. In this article, we provide a quantitative evaluation of the inference capabilities of the different components on mobile SoCs. We also present insights behind their respective power-performance behavior. Finally, we explore the performance limit of the mobile SoCs by synergistically engaging all the components concurrently. We observe that a mobile SoC provides up to 2x improvement with parallel inference when all its components are engaged, as opposed to engaging only one component.