Casper: Carbon-Aware Scalable Processing in Elastic Clusters
The Casper project investigates how the execution of large scalable batch processing applications can be aligned with the availability of low-carbon energy. In particular, we develop methods and prototypes for dynamically managing individual processing steps of scalable cluster applications based on estimates of application performance, resource availability, and carbon intensity.
About Our Research
Our research covers the following aspects:
-
Methods for reducing the emissions of large scalable batch data processing applications on elastic compute clusters
-
Prototypes for elastic Kubernetes clusters and widely used batch processing frameworks (such as Nextflow and Spark)
-
Experiments assessing these methods and prototypes on public- and private-cloud Kubernetes deployments
Latest News
PECS workshop paper on Spark energy estimation presented at Euro-Par 2026
August 25, 2026
We presented our PECS 2026 workshop paper on the accuracy of estimating Spark energy consumption using resource utilisation metrics and power models.
Read moreNew preprint presenting the Ichnos+ workflow footprint estimator
July 16, 2026
A new preprint detailing Ichnos+, a system for estimating the environmental footprint of scientific workflows using fitted power models, is up on arXiv.
Read moreIEEE CLOUD 2026 research paper on workflow energy prediction
June 11, 2026
Our full conference paper on Augur predicts workflow energy consumption before execution and was accepted for IEEE CLOUD 2026.
Read moreRecent Publications and Open-Source Software
How Accurately Can the Energy Use of Spark Applications Be Estimated Based on Resource Utilisation?
Youssef Moawad, Kathleen West, Vasilis Bountris, Philipp Thamm, Yehia Elkhatib, and Lauritz Thamsen
To appear in the Proceedings of the Euro-Par 2026 Workshops (Euro-Par), 2026
Augur: Pre-Execution Energy Prediction for Workflow Tasks in Heterogeneous Clusters
Kathleen West, Vasilis Bountris, Philipp Thamm, Ulf Leser, Yehia Elkhatib, and Lauritz Thamsen
To appear in the Proceedings of the 19th IEEE International Conference on Cloud Computing (CLOUD), 2026
Energy-Aware Workflow Execution: An Overview of Techniques for Saving Energy and Emissions in Scientific Compute Clusters
Lauritz Thamsen, Yehia Elkhatib, Paul Harvey, Syed Waqar Nabi, Jeremy Singer, and Wim Vanderbauwhede
Workflow Systems for Large-Scale Scientific Data Analysis, 2026