Green Algorithms: Towards More Sustainable Scientific Computing

High-performance computing (HPC) is an essential component of modern research, and it comes with significant, but not always well-understood, environmental impacts. The Green Algorithms Initiative promotes more environmentally sustainable computational science by building tools for carbon footprint estimation, raising awareness, and driving change in the computational science community. Led by the Cambridge Sustainable Computing Lab, there are three main arms to the project: research and tools, frameworks and training for sustainable research, and policy engagement. 

The Green Algorithms toolkit offers three tools covering different user profiles. The online calculator is a widely used tool that can be used to quickly estimate the carbon footprint of a computational task without installing anything. GA4HPC and the Green Algorithms Dashboard automate this process for institutional computing facilities. GA4HPC is a command-line tool that individual HPC users run on their own to get a footprint report for their own computational workloads, while the Dashboard offers ongoing, automated reporting of a user’s own usage, plus a big-picture view across their team, department, and institution.

Also: a free on-demand training course was developed with EMBL-EBI on the topic of sustainable computing and available on the EBI training platform. The GREENER framework for sustainable computational science provides guiding principles for the field and has helped funders and policy makers to develop their policies in this space. 

By raising awareness about this issue, and providing resources for scientists to tackle it, the Green Algorithms Initiative helps to ensure that the environmental impacts of the research produced by computational scientists don’t outweigh its benefits to society.