Prof. Dr. Jonas Posner

Professor in Computer Science at the Fulda University of Applied Sciences, Germany.

jonasposner.png

Professor of Operating Systems and Parallel Processing at Fulda University of Applied Sciences since 2025. My research aims to make supercomputers more efficient and easier to program. It focuses on Asynchronous Many-Task (AMT) runtime systems (work-stealing load balancing, task-level fault tolerance) and on dynamic resource management (malleable and evolving programs, malleable and energy-aware job scheduling), complemented by HPC education and AI-assisted parallel programming.

These efforts resulted in 30+ peer-reviewed publications, a DFG individual research grant (€ 380,000), invited talks at LANL and in Kobe (Japan), and service as Technical Program Chair of WAMTA 2027, on the program committees of SC and IPDPS, and on workshop program committees at Euro-Par and PPAM. I lead a research group with a postdoc and a Ph.D. student and, via PZAI, hold the right to supervise doctorates.

My full CV is available here.

News

Nov 19, 2026

Two new preprints on arXiv!

Can Agentic LLMs Write Task-Based Parallel Code? A Study of OpenMP Tasking, HPX, StarPU and TTG: https://arxiv.org/abs/2611.19105

Thinflow: Organizing HPC Benchmarking Campaigns as Rerunnable Directory Trees: https://arxiv.org/abs/2611.26526

Nov 15, 2026

Two papers published in the SC26 Workshops proceedings!

From a Last-Week Add-On to First-Day Practice: Three Years of Integrating MPI Performance Analysis into HPC Education with EduMPI Suite (EduHPC): 10.1109/SCW72768.2026.00190, slides: 10.5281/zenodo.24248845

Saving Energy Costs and Time on Supercomputers: Scheduler-Integrated Recurring Node Shutdown Exploiting Malleable Jobs (EESP): 10.1109/SCW72768.2026.00195, slides: 10.5281/zenodo.24251546

The SC26 Doctoral Showcase poster Near-Real-Time Performance Visualization of MPI Applications: Integrating Performance Analysis into HPC Education is also online: 10.5281/zenodo.23059850

Aug 31, 2026

New paper published! Performance Analysis in Parallel Programming Education: A Comparative Usability Study.

It is now available online at: https://dl.gi.de/handle/20.500.12116/48841

Aug 05, 2026

New paper published! Evaluating Malleable Job Scheduling in HPC Clusters using Real-World Workloads.

It is now available online at: 10.1007/978-3-032-24923-4_1

Jul 16, 2026

New paper published! Exploring Performance-Productivity Trade-offs in AMT Runtimes: A Task Bench Study of Itoyori, ItoyoriFBC, HPX, and MPI.

It is now available online at: 10.1007/978-3-032-27676-6_1

Talks and collaborations

Cyprus: Limassol France: Orleans Germany: Darmstadt, Frankfurt, Garching, Hamburg, Kassel, Stuttgart Jamaica: Kingston Japan: Kobe Poland: Lublin, Poznań Spain: Madrid U.S.: Atlanta, Chicago, Dallas, Denver, Gleneden Beach, Knoxville, Lake Buena Vista, Los Alamos, St. Louis