2023 International Workshop on Performance, Portability & Productivity in HPC


The P3HPC workshop will be held on Monday, 13 November 2023 from 9:00AM - 12:30PM Mountain.

Session 1: Applications

Session Chair: Scott Parker

09:03 - 09:22 MST
A Performance-Portable SYCL Implementation of CRK-HACC for Exascale
Esteban Rangel, John Pennycook, Adrian Pope, Nicholas Frontiere, Zhiqiang Ma, and Varsha Madananth

09:22 - 9:41 MST
Performance Evaluation of Heterogeneous GPU Programming Frameworks for Hemodynamic Simulations
Aristotle Martin, Geng Liu, William Ladd, Seyong Lee, John Gounley, Jeffrey Vetter, Saumil Patel, Silvio Rizzi, Victor Mateevitsi, Joseph Insley, and Amanda Randles

09:41 - 10:00 MST
Performance Portability Evaluation of Blocked Stencil Computations on GPUs
Oscar Antepara, Samuel Williams, Hans Johansen, Tuowen Zhao, Samantha Hirsch, Priya Goyal, and Mary Hall

Morning Break

10:00 - 10:30 MST

Session 2: Applications and Algorithms

Session Chair: John Pennycook

10:30 - 10:39 MST
Benchmarking a portable lattice quantum chromodynamics kernel written in Kokkos and MPI
Simon Schlepphorst and Stefan Krieg

10:39 - 10:48 MST
MatRIS: Multilevel Math Library Abstraction for Heterogeneity and Performance Portability using IRIS Runtime
Mohammad Alaul Haque Monil, Narasinga Rao Miniskar, Keita Teranishi, Jeffrey S. Vetter, and Pedro Valero-Lara

10:48 - 10:57 MST
Porting Batched Iterative Solvers onto Intel GPUs with SYCL
Phuong Nguyen, Pratik Nayak, and Hartwig Anzt

Session 3: Networking

Session Chair: Judy Hill

10:57 - 11:16 MST
Evaluating the Performance of One-sided Communication on CPUs and GPUs
Nan Ding, Muhammad Haseeb, Taylor Groves, and Samuel Williams

11:16 - 11:25 MST
Performance Portability of Programming Strategies for Nearest-Neighbor Communication with GPU-Aware MPI
James B. White III

Session 4: GPUs

Session Chair: CJ Newburn

11:25 - 11:44 MST
Evaluating the performance portability of SYCL across CPUs and GPUs on bandwidth-bound applications
István Z. Reguly

11:44 - 12:03 MST
CuPBoP-AMD: Extending CUDA to AMD Platforms
Jun Chen, Xule Zhou, and Hyesoon Kim

12:03 - 12:12 MST
High-level GPU code: a case study examining JAX and OpenMP
Nestor Demeure, Theodore Kisner, Reijo Keskitalo, Rollin Thomas, Julian Borrill, and Wahid Bhimji

12:12 - 12:21 MST
Many Cores, Many Models: GPU Programming Model vs. Vendor Compatibility Overview
Andreas Herten