The Ecological Footprint of Neural Machine Translation Systems

TL;DR

This paper examines the ecological impact of neural machine translation systems by analyzing their power consumption, carbon footprint, and potential mitigation strategies like quantization across different architectures and hardware.

Contribution

It provides a comprehensive assessment of the environmental footprint of neural MT, comparing architectures, hardware, and proposing quantization as a power-efficient solution.

Findings

Transformers consume more power than RNNs during training.

GPU-based training has a significant carbon footprint, comparable to household appliances.

Quantization reduces power consumption and enables CPU-based inference.

Abstract

Over the past decade, deep learning (DL) has led to significant advancements in various fields of artificial intelligence, including machine translation (MT). These advancements would not be possible without the ever-growing volumes of data and the hardware that allows large DL models to be trained efficiently. Due to the large amount of computing cores as well as dedicated memory, graphics processing units (GPUs) are a more effective hardware solution for training and inference with DL models than central processing units (CPUs). However, the former is very power demanding. The electrical power consumption has economical as well as ecological implications. This chapter focuses on the ecological footprint of neural MT systems. It starts from the power drain during the training of and the inference with neural MT models and moves towards the environment impact, in terms of carbon dioxide emissions. Different architectures (RNN and Transformer) and different GPUs (consumer-grate NVidia 1080Ti and workstation-grade NVidia P100) are compared. Then, the overall CO2 offload is calculated for Ireland and the Netherlands. The NMT models and their ecological impact are compared to common household appliances to draw a more clear picture. The last part of this chapter analyses quantization, a technique for reducing the size and complexity of models, as a way to reduce power consumption. As quantized models can run on CPUs, they present a power-efficient inference solution without depending on a GPU.