[hpc-announce] [2nd CFP] Asynchronous Many-Task Systems for Exascale Workshop (AMTE) @ PPAM 2026

[Jonas Posner]

We are pleased to invite submissions to the Asynchronous Many-Task 
Systems for Exascale Workshop (AMTE), held in conjunction with the 16th 
International Conference on Parallel Processing and Applied Mathematics 
(PPAM 2026).

Workshop website: https://urldefense.us/v3/__https://amte-workshop.github.io__;!!G_uCfscf7eWS!canCrvewvJ5EczVy3F4VTBGmMcIpd7LhoiRI3MQkAU2eiaaHZxNXbeiXMfVJa2Ms9wCfc4xkR0FyZhTsBbCb-myHBssfRxKnmiuTV-4yF54$ 

Conference website: https://urldefense.us/v3/__https://ppam.edu.pl__;!!G_uCfscf7eWS!canCrvewvJ5EczVy3F4VTBGmMcIpd7LhoiRI3MQkAU2eiaaHZxNXbeiXMfVJa2Ms9wCfc4xkR0FyZhTsBbCb-myHBssfRxKnmiuTrrN9__w$ 

Location: Poznań, Poland
Dates: August 30 – September 2, 2026

Abstract:
Supercomputers have begun operating at exascale performance, and a 
tremendous amount of work has been invested in identifying and 
overcoming the challenges leading up to this milestone. These challenges 
include load balancing, high-throughput data movement, and efficient 
resource utilization. Asynchronous Many-Task (AMT) programming models 
and runtime systems have shown that these challenges can be addressed by 
providing mechanisms such as oversubscription, locality-aware 
scheduling, shared-memory execution, and data-dependence–driven execution.

The Asynchronous Many-Task systems for Exascale Workshop (AMTE) explores 
the advantages of AMT programming on current and emerging HPC systems. 
It will gather developers, users, and researchers to share experiences, 
discuss how their approaches meet the challenges posed by today’s 
heterogeneous exascale architectures, and explore opportunities for 
increased performance, robustness, portability, and full-system utilization.

Topics of Interest (but not limited to):
- Novel AMT runtime systems
- Experiences using AMT runtime systems
- Comparisons between AMT runtime systems
- Task coordination mechanisms (e.g., dataflow, fork–join)
- Using AMT runtime systems for accelerators or heterogeneous architectures
- Benchmarks for AMT runtime systems
- Profiling, performance analysis, and debugging tools for AMT runtime 
systems
- Challenges in scaling AMT runtime systems to exascale supercomputer
- Hardware challenges and co-design opportunities for AMT runtime systems
- Task-based algorithms and applications
- AMT approaches for emerging domains beyond HPC (e.g., IoT, distributed 
systems)
- Application-specific optimizations and case studies using AMT runtime 
systems

Important Dates:
- Submission deadline: April 24, 2026 (AOE)
- Acceptance notification: May 25, 2026 (AOE)
- Workshop: August 30 – September 2, 2026

The workshop will be held for one or two days. Papers must be original 
and not previously published or under review elsewhere. Full papers may 
not exceed 15 pages (LNCS format). Extended abstracts of up to 4 pages 
are also welcome. Only accepted full papers will be published in the 
proceedings.

The workshop program will include a keynote talk, an invited talk, and 
presentations of accepted full papers and extended abstracts.

Submissions must be made via the PPAM 2026 EasyChair system:
https://urldefense.us/v3/__https://easychair.org/my/conference?conf=ppam2026__;!!G_uCfscf7eWS!canCrvewvJ5EczVy3F4VTBGmMcIpd7LhoiRI3MQkAU2eiaaHZxNXbeiXMfVJa2Ms9wCfc4xkR0FyZhTsBbCb-myHBssfRxKnmiuT8Dl-Q2w$ 


Sincerely,
Patrick Diehl, Los Alamos National Laboratory, USA
Jonas Posner, Fulda University of Applied Sciences, Germany
Claudia Fohry, University of Kassel, Germany
Parsa Amini, Advanced Micro Devices, Inc., USA