Casper: Carbon-Aware Scalable Processing in Elastic Clusters

IEEE CLOUD 2026 research paper on workflow energy prediction

June 11, 2026

Our full conference paper, “Augur: Pre-Execution Energy Prediction for Workflow Tasks in Heterogeneous Clusters” was accepted for publication in the proceedings of the “2026 IEEE International Conference on Cloud Computing (CLOUD), and will be presented in Sydney in July 2026.

In the paper, we proposed Augur, a novel method to predict the energy consumption of scientific workflow tasks prior to execution. By efficiently profiling both the available cluster infrastructure and the workflow at hand, Augur is capable of predicting the overall energy consumption with a median prediction error of 16.3% compared to Ichnos, an energy estimation method that uses fitted power models, and 18.2% compared to Intel RAPL measurements, as observed in our experimental evaluation on public and private cloud infrastructure. Augur outperforms two state-of-the-art methods in predicting both task runtime and total workflow energy, providing a robust foundation for energy-efficient and carbon-aware scientific data analysis.

The pre-print of the paper is available on arXiv, with the results repository also available on GitHub.

Future Generation Computer Systems article on carbon-aware scientific workflows

March 06, 2026

We are happy to share that our journal article “A Systematic Evaluation of the Potential of Carbon-Aware Execution for Scientific Workflows” has been published in the journal Future Generation Computer Systems.

FONDA workshop presentations on workflow carbon impact at HU Berlin

November 06, 2025

Kathleen and Lauritz will be speaking at the FONDA workshop on Reducing the Carbon Emissions of Large-Scale Computational Data Analysis, which will take place on 25 and 26 November at Humboldt University of Berlin.