We are excited to announce the official start of the EPSRC-funded Casper project, aiming to advance carbon-aware computing to tackle the challenge of aligning flexible cloud computing workloads with low-carbon energy availability.

ICT infrastructure accounts for 2-3% of global energy consumption, rivalling commercial aviation, with cloud data centres contributing around a third of this. Casper aims to reduce emissions from large-scale batch data processing applications in elastic cloud environments by dynamically scheduling and scaling workloads when renewable energy is abundant. Through the development of the first Stepwise Performance-, Interruption-, Resource, Carbon-Aware Scheduler (SPIRCS), we aim to enable more sustainable batch data processing on cloud infrastructure.

Collaborating with our partners, AWS, the BBC, and HU Berlin, Casper will be actively working towards real-world demonstrations and deployments to deliver impact on the idea of carbon-aware cloud-based batch processing.

Stay tuned to find out more about our research, and in the meantime, get to know our wider research team and have a look at our very first outputs on this website.