School of Computing Science

Abstract: As federated learning (FL) extends from distributed machine learning between low-power devices to cross-silo scenarios involving edge servers and data centres, its carbon footprint has become a growing concern. Addressing this, methods for sustainable FL align training with low-carbon energy availability or low grid demand and reduce the energy consumption of clients powered by high-carbon sources by decreasing the size of their models.

We propose applying model partitioning, which can shift energy consumption by offloading parts of a model to another participant, in response to carbon- or grid-aware signals. Our preliminary findings show that for some partition points, model partitioning can reduce a participant’s energy consumption by up to 76% without any significant time or energy consumption overhead compared to non-partitioned training.

 

Title: Scoring and Gamification to Encourage Sustainable Use of Compute Clusters
Abstract: Computing's environmental footprint keeps growing, but behaviour change lags behind. We introduce a composite sustainability score (0–100) designed for gamified feedback, combining carbon intensity, resource utilisation, and embodied emissions to reward carbon-aware scheduling, efficient utilisation, and use of underutilised cluster hardware. This scoring system is built into the IDA Launcher cluster management interface and underpins three dashboard conditions: raw metrics, composite score, and a gamified tree visualisation, which we are planning to evaluate in a 12-week within-subjects study with approximately 35 researchers.
We present this scoring system, the planned evaluation and discuss the challenge of defining computational work 'goodness' in the context of sustainability scores.