Prose

Tobias Weinzierl is Professor of High-Performance Computing (HPC) in the Department of Computer Science at Durham University, where he leads the Scientific Computing research group and serves as the Director of Durham’s Institute for Data Science (IDAS). After studying Computer Science with a minor in Maths, he earned a Dr. rer. nat. (PhD) as well as a habilitation in Computer Science from TUM. Tobias has been the inaugurate director of the Master in Scientific Computing and Data Analysis (MISCADA), he is the PI on multiple HPC projects tied to the UK’s exascale programme ExCALIBUR, and he heads the UK’s first Intel oneAPI Centre of Excellence. He also serves as Domain Panel Chair (Artificial Intelligence) for EuroHPC JU.
He is particularly interested in efficient ways how to translate state-of-the-art algorithms – multigrid, higher-order DG or SPH formalisms – into fast code that fits to modern architectures, and how to create performance-portable algorithms and code. Where possible, his work feeds into open source software. This software has to challenge modern hardware and software stacks, including compilers, tools, and runtimes, while demonstrating how we can obtain better insights more efficiently from modern technology by exploiting systems creatively and employing the latest numerical schemes.
In his role as Director of IDAS, Tobias is particularly interested in upskilling Digital Research Technology Professionals (RTPs). He champions several large grants within the UK’s Digital Research Infrastructure initiative that focus on skills development (including CAKE, SHAREing, and HAI-End), and is therefore heavily involved in creating advanced upskilling materials for HPC, developing novel training formats, and establishing a productive and efficient HPC software and service landscape.

- Another great @durham-comp-sci.bsky.social talk by Chung on his UQ work and load balancing at #PPAM2026
- My @cake-dri.bsky.social colleague @nickbrown-hpc.bsky.social gives an enthusiastic talk on RISC-V. Nice to see and hear about the different pathways of ARM and RISC-V.
- Mladen presents our work on SPH offloading to GPUs at #PPAM2026. Slight hickups with the mic, but then really exciting talk.
- On my way to Poznan for #PPAM2026. This conference is always a highlight of the year. Looking forward to meet friends and learn about the cutting edge in scientific computing. Also an honour to contribute one of the keynotes: ppam.edu.pl https://ppam.edu.pl/call
- Paper alert: thanks to funding through the #eCSE programme we have completed a first version of our domain specific language DSLHyPE for ExaHyPE to model compute kernels for hyperbolic PDEs for CPUs and GPUs. This is joint work with friends from @epcc.bsky.social : arxiv.org/abs/2608.19273 https://arxiv.org/abs/2608.19273