Eco-conception

The question of how to incorporate sustainability into our lives has long been debated. Can we integrate it at the beginning, before it…

This article attempts to estimate the carbon footprint of the popular OpenAI chatbot called ChatGPT

The web is a major source of carbon emissions. This site provides independent research to help web developers make their products more sustainable.

Want to be true in advertising, earn trust and avoid lawsuits? Steer clear of this lingo (and use other terms with care).

Green Service Design - Pour une conception de services éco-responsables, le 1er livre dédié à l'éco-conception de services

Conception et production du nouveau site web de Nature en Occitanie

The computation demand for machine learning (ML) has grown rapidly recently, which comes with a number of costs. Estimating the energy cost helps measure its environmental impact and finding greener strategies, yet it is challenging without detailed information. We calculate the energy use and carbon footprint of several recent large models-T5, Meena, GShard, Switch Transformer, and GPT-3-and refine earlier estimates for the neural architecture search that found Evolved Transformer. We highlight the following opportunities to improve energy efficiency and CO2 equivalent emissions (CO2e): Large but sparsely activated DNNs can consume 1/10th the energy of large, dense DNNs without sacrificing accuracy despite using as many or even more parameters. Geographic location matters for ML workload scheduling since the fraction of carbon-free energy and resulting CO2e vary ~5X-10X, even within the same country and the same organization. We are now optimizing where and when large models are trained. Specific datacenter infrastructure matters, as Cloud datacenters can be ~1.4-2X more energy efficient than typical datacenters, and the ML-oriented accelerators inside them can be ~2-5X more effective than off-the-shelf systems. Remarkably, the choice of DNN, datacenter, and processor can reduce the carbon footprint up to ~100-1000X. These large factors also make retroactive estimates of energy cost difficult. To avoid miscalculations, we believe ML papers requiring large computational resources should make energy consumption and CO2e explicit when practical. We are working to be more transparent about energy use and CO2e in our future research. To help reduce the carbon footprint of ML, we believe energy usage and CO2e should be a key metric in evaluating models, and we are collaborating with MLPerf developers to include energy usage during training and inference in this industry standard benchmark.

The main training portal of the GSF. Contribute to Green-Software-Foundation/training development by creating an account on GitHub.

“Le conseil #LogicielsLibres, mis en place et animé par la mission logiciels libres de la @_DINUM à @Etalab, émet son premier avis en répondant positivement à cette question:”

If you want to make your software applications more sustainable and build a career in green software, then this training is for you

Learn the basic concepts a software practitioner needs to know to build, maintain and run greener applications.

Accenture shares the different levels of ambition that organizations can expect as they begin their journey toward a sustainable cloud. Read more.

MiNumEco, la mission interministérielle numérique écoresponsable pilotée par la DINUM et le Ministère de la Transition écologique pour réduire les impacts environnementaux du numérique

We outline in more detail the process of listing providers as green, but if you are not sure about what to present as evidence we have outlined what we accept here, and our reasoning. The general principle is as follows though: If you want a green site, you need to demonstrate steps you are taking… Read More »What we accept as evidence of green power

An open source tool to measure and analyze cloud carbon emissions

Dear diary, We’ve been asked by Chris and Michelle to write a post that summarises 5 things we’ve learned in the fellowship so far. It’s been a great prompt to take stock and summarise the vast amount of learning I’ve done. I dashed off a quick list a few weeks back. Since then, whenever something… Read More »What people think building a sustainable internet involves vs what it might actually be

SDIA is the first alliance created as an authoritative source of industry leaders and information forging the way for sustainable digital infrastructure.

CEDaCI increases collaboration and communication to drive sustainability in the data industry.

Sign our pledge and support the development of an independent Type I sustainability certificate that helps you procure responsibly. Identify suppliers who meet strong environmental and social requirements, and who deliver the transparent information required for the certificate's sustainability reporting.

At WCEU 2022, Nora Ferreirós from the Spanish WordPress community, asked Matt and Josepha about their thoughts on supporting sustainability concerns within the community. Matt suggested setting up …

null

PowerJoular allows monitoring power consumption of multiple platforms and processes. - GitHub - joular/powerjoular: PowerJoular allows monitoring power consumption of multiple platforms and processes.

A database containing power models for various computing components, and hardware devices. - joular/powermodels: A database containing power models for various computing components, and hardware de...