Markus Götz

Markus Götz

Assistant Professor for Scalable and Efficient Machine Learning

Publications

A full list of articles is on Google Scholar.

Journal Articles

The PAINT database for operational concentrating solar power plant data following FAIR data principles

Published in Nature Energy, 2026

We introduce PAINT, the first FAIR, open-access database of operational concentrating solar power tower plant data, providing 849 GB of heliostat, calibration, and weather data to accelerate AI-driven solar research.

Recommended citation: Phipps, K., Kuhl, M., Weiel, M., Busch, M., Lewen, J., Blumenröhr, N., Maldonado Quinto, D., Debus, C., Göhring, F., Kaufhold, O., Streit, A., Pitz-Paal, R., Götz, M., & Pargmann, M. (2026). "The PAINT database for operational concentrating solar power plant data following FAIR data principles." Nature Energy.
Download Paper

Deep learning for augmented process monitoring of scalable perovskite thin-film fabrication

Published in Energy & Environmental Science, 2025

We combine deep learning with in-situ luminescence imaging to monitor and better understand the scalable fabrication process of perovskite thin films, enabling augmented, data-driven process control for photovoltaics manufacturing.

Recommended citation: Laufer, F., Götz, M., & Paetzold, U. W. (2025). "Deep learning for augmented process monitoring of scalable perovskite thin-film fabrication." Energy & Environmental Science.
Download Paper

Automatic heliostat learning for in situ concentrating solar power plant metrology with differentiable ray tracing

Published in Nature Communications, 15(1), 6997, 2024

We use differentiable ray tracing to automatically calibrate heliostats in situ, learning precise geometric alignment metrology for concentrating solar power plants directly from operational sensor data.

Recommended citation: Pargmann, M., Ebert, J., Götz, M., Maldonado Quinto, D., Pitz-Paal, R., & Kesselheim, S. (2024). "Automatic heliostat learning for in situ concentrating solar power plant metrology with differentiable ray tracing." Nature Communications, 15(1), 6997.
Download Paper

Conference Papers

Combining Domain and Tensor Parallelism to Train Multi-billion-parameter AI Weather Models

Published in Proceedings of the Platform for Advanced Scientific Computing Conference (PASC 26), 2026

We introduce Jigsaw, a model parallelization scheme combining domain and tensor parallelism to train multi-billion-parameter AI weather models, eliminating memory redundancy and achieving state-of-the-art scaling on up to 256 GPUs.

Recommended citation: Kieckhefen, D., Götz, M., Heyen, L. H., Streit, A., & Debus, C. (2026). "Combining Domain and Tensor Parallelism to Train Multi-billion-parameter AI Weather Models." In: Proceedings of the Platform for Advanced Scientific Computing Conference (PASC 26).
Download Paper

perun: Benchmarking Energy Consumption of High-Performance Computing Applications

Published in European Conference on Parallel Processing (Euro-Par), Springer, pp. 17–31, 2023

We present perun, an open-source tool for measuring and benchmarking the energy consumption of high-performance computing applications, enabling more transparent and sustainable HPC and AI research.

Recommended citation: Gutiérrez Hermosillo Muriedas, J. P., Flügel, K., Debus, C., Obermaier, H., Streit, A., & Götz, M. (2023). "perun: Benchmarking Energy Consumption of High-Performance Computing Applications." In: European Conference on Parallel Processing. Springer, pp. 17–31.
Download Paper

HPDBSCAN: Highly Parallel DBSCAN

Published in Proceedings of the Workshop on Machine Learning in High-Performance Computing Environments (MLHPC 15), pp. 1–10, 2015

We present HPDBSCAN, a highly parallel, distributed-memory implementation of the DBSCAN clustering algorithm that scales density-based clustering to large datasets on high-performance computing systems.

Recommended citation: Götz, M., Bodenstein, C., & Riedel, M. (2015). "HPDBSCAN: Highly Parallel DBSCAN." In: Proceedings of the Workshop on Machine Learning in High-Performance Computing Environments (MLHPC 15), pp. 1–10.
Download Paper