Program
The P3HPC workshop will be held on Monday, 18 November 2024 from 9:00AM - 5:30PM Eastern.
Session 1
09:05 - 09:45 EST
Featured Speaker
Julien Bigot
09:45 - 10:00 MST
Optimizing MILC-Dslash Performance on NVIDIA A100 GPU: Parallel Strategies using SYCL
Amanda S. Dufek, Steven A. Gottlieb, Muaaz Gul Awan, Douglas Adriano Augusto, Jack Deslippe, and Brandon Cook
Morning Break
10:00 - 10:30 EST
Session 2
10:30 - 10:55 EST
A Metric for HPC Programming Model Productivity
Tom Lin, Tom Deakin, and Simon McIntosh-Smith
10:55 - 11:20 EST
High-Performance, Scalable Geometric Multigrid via Fine-Grain Data Blocking for GPUs
Oscar Antepara, Samuel Williams, Hans Johansen, and Mary Hall
11:20 - 11:45 EST
Autonomous Execution for Multi-GPU Systems: Compiler Support
Javid Baydamirli, Tal Ben Nun, and Didem Unat
11:45 - 12:10 EST
Performance and Power: Systematic Evaluation of AI Workloads on Accelerators with CARAML
Chelsea Maria John, Andreas Herten, Stepan Nassyr, and Carolin Penke
12:10 - 12:25 EST
Development of Performance Portable Spline Solver for Exa-Scale Plasma Turbulence Simulation
Yuuichi Asahi, Baptiste Legouix, Emily Bourne, Thomas Padioleau, Julien Bigot, Virginie Grandgirard, and Kevin Obrejan
Lunch Break
12:30 - 14:00 EST
Session 3
14:00 - 14:23 EST
Performance Modeling and Analysis of a de Bruijn Graph Based Local Assembly Kernel on Multiple Vendor GPUs
LeAnn Lindsey, Nan Ding, Jack DeSlippe, and Muaaz Awan
14:23 - 14:46 EST
Performance Portability via C++ PSTL, SYCL, OpenMP, and HIP: The Gaia AVU-GSR Case Study
Giulio Malenza, Valentina Cesare, Marco Edoardo Santimaria, Robert Birke, Alberto Vecchiato, Ugo Becciani, and Marco Aldinucci
14:46 - 15:00 EST
Performance Portable Optimizations of an Ice-sheet Modeling Code on GPU-supercomputers
Oscar Antepara, Samuel Williams, Max Carlson, and Jerry Watkins
Afternoon Break
15:00 - 15:30 EST
Session 4
15:30 - 15:55 EST
RAJA Performance Suite: Performance Portability Analysis with Caliper and Thicket
Olga Pearce, Jason Burmark, Rich Hornung, Befikir Bogale, Ian Lumsden, Michael McKinsey, Dewi Yokelson, David Boehme, Stephanie Brink, Michela Taufer, and Tom Scogland
Lightning Talks
15:55 - 16:01 EST
An Analysis into the Performance and Productivity of Rust in High Performance Computing
Edmund Goodman, Richard Kirk
16:01 - 16:07 EST
Performance and Scaling of HPC and AI Applications on Leadership Class Intel, AMD, and NVIDIA GPU-Accelerated Systems
JaeHyuk Kwack, Colleen Bertoni, Umesh Unnikrishnan, Riccardo Balin, Khalid Hossain, Yasaman Ghadar, Timothy Williams, Abhishek Bagusetty, Mathialakan Thavappiragasam, Väinö Hatanpää, Archit Vasan, John Tramm, and Scott Parker
16:07 - 16:13 EST
Exploring SYCL for Batched Kernels with Memory Allocations
Aymeric Millan, Thomas Padioleau, and Julien Bigot
16:13 - 16:19 EST
Evaluating Performance Portability of a Seismic Survey Simulation across GPU Architectures
Arthur Lorenzon, Philippe Navaux, Alexandre Sardinha, and Bronson Messer
16:19 - 16:25 EST
Leveraging AI to port from legacy Fortran to GPU enabled C++
Hannah Elizabeth Ross, and Jean Sexton
16:25 - 16:40 EST
Lightning Talk Q&A
Panel
16:40 - 17:25 EST
Tooling Panel & Discussion
Moderators:Doug Jacobsen and CJ Newburn
Panelists:Kaan Olgu, Tom Lin, Solomon Bekele