MENU

Open-source tool tracks AI computing’s CO2 footprint

Open-source tool tracks AI computing’s CO2 footprint

New Products |
By Rich Pell



The software, called CodeCarbon, is a lightweight software package that seamlessly integrates into Python codebase. It estimates the amount of carbon dioxide (CO2) produced by the computing resources used to execute the code to incentivize developers to optimize their code efficiency.

It also advises developers on how they can reduce emissions by selecting their cloud infrastructure in regions that use lower carbon energy sources, say the software’s developers.

“AI is a powerful technology and a force for good, but it’s important to be conscious of its growing environmental impact,” says Yoshua Bengio, Mila founder and Turing Prize recipient. “The CodeCarbon project aims to do just that, and I hope that it will inspire the AI community to calculate, disclose, and reduce its carbon footprint.”

Sylvain Duranton, a managing director and senior partner at Boston Consulting Group (BCG) and global head of BCG GAMMA, adds, “If recent history is any indicator, the use of computing in general, and AI computing in particular, will continue to expand exponentially around the world. As this happens, CodeCarbon can help organizations make sure their collective carbon footprint increases as little as possible.”

The software’s developers say such a tool is needed to help address the amount of energy needed to support the massive computing behind AI. For example, training a powerful machine-learning algorithm can require running multiple computing machines for days or weeks.

And, say the developers, the fine-tuning required to improve an algorithm by searching through different parameters can be especially intensive. For recent state-of-the-art architectures like VGG, BERT, and GPT-3, which have millions of parameters and are trained on multiple GPUs (graphic processing units) for several weeks, this can mean a difference of hundreds of kilograms of CO2eq – a standardized measure used to express the global warming potential of various greenhouse gases.

The software’s emission tracker records the amount of power being used by the underlying infrastructure from major cloud providers and privately hosted on-premise datacenters. Based on publicly available data sources, it estimates the amount of CO2 emissions produced by referring to the carbon intensity from the energy mix of the electric grid to which the hardware is connected.

The tracker logs the estimated CO2 equivalent produced by each experiment and stores the emissions across projects and at an organizational level. This gives developers greater visibility into the amount of emissions generated from training their models and makes the amount of emissions tangible in a user-friendly dashboard by showing equivalents in easily understood numbers like automobile miles driven, hours of TV watched, and daily energy consumed by an average U.S. household.

The ability to track CO2 emissions, say the developers, represents a significant step forward in developers’ ability to use energy resources wisely and, therefore, reduce the impact of their work on an increasingly fragile environment. CodeCarbon is also expected to help introduce greater transparency into the developer community, enabling developers to measure and then report emissions created by an array of computing experiments.

Jonathan Wilson, Associate Professor of Environmental Studies at Haverford College, says, “Computing’s carbon footprint depends on where the computations are performed, how much power is consumed, and whether fossil fuels or low-carbon sources generate that electricity. CodeCarbon will show you where to run your code to minimize your carbon footprint.”

Niko Laskaris, data scientist, Comet.ml, adds, “Our community needs to innovate more responsibly, and that starts with tracking and optimizing your model. With CodeCarbon, data scientists and teams can keep building great models, but with a new parameter: the carbon footprint of their work.”

The CodeCarbon developers also say they look forward to developers and researchers using the tool and contributing to it by enhancing it with new capabilities. To increase awareness of the environmental impact of computing, they recommend that users report the CO2eq of their experiments in research papers, articles, and tech blogs.

Mila
BCG GAMMA
Comet.ml

Related articles:
Optical processor speeds compute for next-gen AI
Quantum research to study perfect engines, entanglement as fuel
AI bioreactor to bring carbon capture to urban environments
‘Blinking’ nanocrystals may benefit CO2 reduction, quantum computing
Photonic chip slashes energy needed to compute with light

 

If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :    eeNews on Google News

Share:

Linked Articles
10s