Call for Participation: Parallel Applications Workshop, Alternatives To MPI+X (PAW-ATM25, colocated with SC25)
Dan Bonachea
Workshop website: https://go.lbl.gov/paw
Alternatives to MPI+X are worth exploring as programmer productivity becomes a major component of the time to science. Alternatives include parallel programming languages (e.g., Chapel, Regent, Fortran 2023), general purpose libraries (e.g., Charm++, COMPSs, HPX, Legion, UPC++), and domain specific libraries (e.g., Arkouda, Dask, Spark). With many options to choose from, it is hard for programmers to know which alternative models are appropriate for their application and for programming model developers to understand the opportunities for improvement.
Through discussion of specific applications, PAW-ATM brings together application experts and programming model developers to improve both applications and models.
WORKSHOP PROGRAM
9:00 - 9:05 PAW-ATM2025 Introduction
Karla V. Morris Wright, Elliott Slaughter, Engin Kayraklioglu, Irene Moulitsas, Katherine Rasmussen, Daniele Lezzi and Kenjiro Taura
Session Chair: Katherine Rasmussen - Lawrence Berkeley National Laboratory
9:05 - 10:00 Distinguished Speaker
Pavan Balaji – Meta
10:00 - 10:30 Morning Break (30 min)
Session Chair: Damian W. I. Rouson - Lawrence Berkeley National Laboratory
10:30 - 10:57 Alternative parallel programming models: past, present and is there a future?
Invited Speaker: Laxmikant Kale - University of Illinois at Urbana-Champaign
10:57 - 11:10 User Experience Abstract: A parallel implementation of the immersed boundary method in the Chapel programming language
Willian Carlos Lesinhovski and Nelson Luis Dias
11:10 - 11:30 KVMSR+UDWeave: Extreme-Scaling with Fine-grained Parallelism on the UpDown Graph Supercomputer
Alexander Fell, Yuqing Wang, Tianshuo Su, Marziyeh Nourian, Wenyi Wang, Jose M. Monsalve-Diaz, Andronicus Rajasukumar, Jiya Su, Ruiqi Xu, Rajat Khandelwal, Tianchi Zhang, David F. Gleich, Yanjing Li, Hank Hoffmann, and Andrew A. Chien
11:30 - 11:50 Comparing Distributed-Memory Programming Frameworks with Radix Sort
Matt Drozt, Michael P. Ferguson, Ryan D. Friese, and Shreyas Khandekar
11:50 - 12:10 Slicing Is All You Need: Towards A Universal One-Sided Algorithm for Distributed Matrix Multiplication
Benjamin Brock, and Renato Golin
12:10 - 12:30 DiOMP-Offloading: Toward Portable Distributed Heterogeneous OpenMP
Baodi Shan, Mauricio Araya-Polo, and Barbara Chapman
12:30 - 2:00 Lunch Break (90 min)
Session Chair: Jan Ciesko - Sandia National Laboratories
2:00 - 2:20 From Parallel Clusters to Hyper-distributed applications: Programming Swarms with COLMENA
Invited Speaker: Francesc Lordan - Barcelona Supercomputing Center
2:20 - 2:40 Weak Scaling of NVSHMEM Applied To Hashed Distributed Structured Data
Andrew Davis, Hans Johansen, Xinfeng Gao, and Stephen Guzik
2:40 - 3:00 Redesigning GROMACS Halo Exchange: Improving Strong Scaling with GPU-initiated NVSHMEM
Mahesh Doijade, Andrey Alekseenko, Ania Brown, Alan Gray, and Szilárd Páll
3:00 - 3:30 Afternoon Break (30 min)
Session Chair: Francesc Lordan - Barcelona Supercomputing Center
3:30 - 3:50 Enhancing HPX with FleCSI: Automatic Detection of Implicit Task Dependencies
Davis Herring, Maxim Moraru, Scott Pakin, Julien Loiseau, Richard Berger, Philipp V. F. Edelmann, and Ben Bergen
3:50 - 4:10 Stackless vs. Stackful Coroutines: A Comparative Study for RDMA-based Asynchronous Many-Task (AMT) Runtimes
Mia Reitz and Jonas Posner
4:10 - 4:30 KDRSolvers: Scalable, Flexible, Task-Oriented Krylov Solvers
David K Zhang, Rohan Yadav, Alex Aiken, Fredrik Kjolstad, and Sean Treichler
4:30 - 5:30 Panel Discussion: The Role of Alternatives to MPI+X Technologies in AI/ML
Panel Chair: Anshu Dubey - Argonne National Laboratory
Panelists:
Ryan Coffee - SLAC National Accelerator Laboratory
Zhihao Jia - Carnegie Mellon University
Peter Mendygral - Hewlett Packard Enterprise
Jeremy Wilke - NVIDIA
https://sourceryinstitute.github.io/PAW
Best Regards,
PAW-ATM2025 Organizing Committee