February 16th to 18th, 2026
Workshop on Asynchronous Many-Task Systems and Applications 2026
Garching near Munich, Germany
, by , under Creative Commons license.
The venue for all talks is at the Max Planck Computing and Data Facility, on Research Campus Garching, Gießenbachstrasse 2 85748 Garching near Munich, Germany. ** Access via Isarstrasse **. Lecture hall on ground level next to the main entrance to building D2.
Social Events:
- Poster Reception on Monday evening in the showroom of the Max Planck Institute of Plasmaphysics
- Conference Dinner on Tuesday evening in the oldest brewery of the planet, Weihenstephan
Scientific Program and Timetable:
| Time | Mon Feb 16 (Day 1) | Tue Feb 17 (Day 2) | Wed Feb 18 (Day 3) |
|---|---|---|---|
| 08:00 AM | Registration / Coffee and Pastries | Registration / Coffee and Pastries | Registration / Coffee and Pastries |
| 08:50 AM | Welcoming Remarks (Erwin Laure) | ||
| 09:00 AM | Keynote 1, Chair: E. Laure Michael Klemm: Fortran and the OpenMP API in a Modern GPU World |
Keynote 2, Chair: M. Schulz Rosa Badia: Task-based programming models: tradeoffs between granularity and computing platforms |
Keynote 3, Chair: C. Junghans Nick Brown: Asynchronous tasks: the cornerstone of targeting HPC on emerging architectures |
| 10:00 AM | Coffee | Coffee | Coffee |
| Session 1, Chair: M. Rampp | Session 3, Chair: J. Posner | Session 5, Chair: C. Junghans | |
| 10:30 AM | Peter Thoman: Concurrent Scheduling of High-Level Parallel Programs on Multi-GPU Systems | Dominik Huber: Towards Dynamic Resource Management with Charm++ | Vanderlei Munhoz Pereira Filho: Performance Analysis of Task-Based Parallelism for 3D Acoustic Full Waveform Inversion on HPC Infrastructure |
| 11:00 AM | Alexander Strack: Accelerating GPRat: Asynchronous, Task-Based GPU Acceleration for Gaussian Process Regression with HPX | Rüdiger Nather: Finding Conservation Laws of Large Dynamical Systems with Tasks and Futures: A Case Study in Utilizing Dynamic Data Dependencies | Nilesh Chaturvedi: Multiresolution Analysis using Data-Flow Programming |
| 11:30 AM | Vincent Arcila: Hierarchical Scheduling for Composable Affinity-Aware Policies in Task- Based Runtimes | Joseph Schuchart: Using C++ Generators in Dataflow Broadcast | Nicolas Ducarton: Abstract communicators for task based runtime systems |
| 12:00 PM | Lunch | Lunch | Lunch |
| Session 2, Chair: C. Junghans | Session 4, Chair: M. Rampp | Session 6, Chair: M. Rampp | |
| 01:00 PM | Jose Gracia: Evaluating MPI and OpenMP for global task dependencies across processes | Linus Batel: From Prompts to Performance: Evaluating LLMs for Task-based Parallel Code Generation | Zhuowei Gu: Merging Parameterized Task Graphs in PaRSEC through Recursive JDF Composition |
| 01:30 PM | Jonas Posner: Exploring Performance-Productivity Trade-offs in AMT Runtimes: A Task Bench Study of Itoyori, and MPI | Lionel Eyraud-Dubois: First Experiments to Evaluate the Relevance of Task-based Runtime Systems to Implement Large Language Model Applications | Qiao Zhang: Leveraging PaRSEC for Task-Based Heterogeneous Computing in Python and Julia |
| 02:00 PM | Nicolas Brieuc: Managing Explicit Communications Within a Sequential Task Flow Paradigm | Mia Reitz: Silent Data Corruption Protection through Efficient Task Replication | Closing remarks |
| 02:30 PM | Coffee | Coffee | Coffee |
| 03:00 PM | Quantum Computer Tour (until 4:30 PM) | Panel: Novel architectures for AMT, Chair: C. Junghans (until 3:45 PM) | |
| Evening | Reception (05:30 PM - 08:00 PM) |
05:30 PM: Bus to Brewery 06:00 PM - 09:00 PM: Tour / Banquet 09:00 PM: Bus to conference venue |
Contact
If you have any problems or questions, please contact us via e-mail at: diehlpk@lanl.gov