Estimating the Power Consumption of Heterogeneous Devices when performing AI Inference

TL;DR

This paper analyzes the power consumption of the NVIDIA Jetson Nano during AI inference, specifically for object classification with YOLOv5 models, highlighting efficiency differences among variants.

Contribution

It provides an empirical power consumption profile for Jetson Nano performing object detection with YOLOv5, including detailed analysis of throughput and energy efficiency.

Findings

YOLOv5n achieves 12.34 fps and low power consumption of 0.154 mWh/frame

Power consumption varies significantly among YOLOv5 variants

The study offers insights into energy-efficient AI inference on IoT devices

Abstract

Modern-day life is driven by electronic devices connected to the internet. The emerging research field of the Internet-of-Things (IoT) has become popular, just as there has been a steady increase in the number of connected devices. Since many of these devices are utilised to perform CV tasks, it is essential to understand their power consumption against performance. We report the power consumption profile and analysis of the NVIDIA Jetson Nano board while performing object classification. The authors present an extensive analysis regarding power consumption per frame and the output in frames per second using YOLOv5 models. The results show that the YOLOv5n outperforms other YOLOV5 variants in terms of throughput (i.e. 12.34 fps) and low power consumption (i.e. 0.154 mWh/frame).