Workshop: 2022 International Workshop on Performance Portability and Productivity (P3HPC)
Event TypeWorkshop
Registration Categories
W
Session Formats
Recorded
TimeSunday, 13 November 20228:30am - 5pm CST
LocationC155
DescriptionThe aim of this workshop is to bring together developers and researchers with an interest in practical solutions, technologies, tools and methodologies that enable the development of performance-portable applications across a diverse set of current and future high‑performance computers.

We draw from a broad research audience that includes standard languages and runtimes, algorithmic techniques, tools, libraries and non‑standard techniques such as domain-specific languages. We also expect to see submissions from application teams documenting their experiences, good and bad, and those developing metrics and measurement techniques for performance portability and productivity.

This workshop has a proven track record of playing a shepherding role for the broader community in identifying and adapting to technology trends while fostering increased transparency and rigor for performance, portability and productivity. The importance of this role will only increase with the growing diversity and scope of hardware and software platforms, including more-complex workflows and data center scale.

Workshop Website
Presentations
8:30am - 9:15am CSTA Decade of Performance Portability: Lessons Learned
9:15am - 9:37am CSTHeterogeneous Programming for the Homogeneous Majority
Performance Portability
9:37am - 10:00am CSTLeveraging Compiler-Based Translation to Evaluate a Diversity of Exascale Platforms
Performance Portability
10:00am - 10:30am CSTP3HPC – Morning Break
10:30am - 10:52am CSTUnderstanding Strong Scaling on GPUs Using Empirical Performance Saturation Size
Performance Portability
10:52am - 11:15am CSTPortable and Efficient Dense Linear Algebra in the Beginning of the Exascale Era
Performance Portability
11:15am - 11:37am CSTExploiting Dynamic Sparse Matrices for Performance Portable Linear Algebra Operations
Performance Portability
11:37am - 12:00pm CSTPerformance Portability of Sparse Block Diagonal Matrix Multiple Vector Multiplications on GPUs
Performance Portability
12:00pm - 1:30pm CSTP3HPC – Lunch Break
1:30pm - 1:53pm CSTPerformance Portable Vlasov Code with C++ Parallel Algorithm
Performance Portability
1:53pm - 2:15pm CSTToward Cross-Platform Portability of Coupled-Cluster Methods with Perturbative Triples Using SYCL
Performance Portability
2:15pm - 2:38pm CSTFrom Task-Based GPU Work Aggregation to Stellar Mergers: Turning Fine-Grained CPU Tasks into Portable GPU Kernels
Performance Portability
2:38pm - 3:00pm CSTPiper: Pipelining OpenMP Offloading Execution through Compiler Optimization for Performance
Performance Portability
3:00pm - 3:30pm CSTP3HPC – Afternoon Break
3:30pm - 3:53pm CSTToward Performance Portability of AI Graphs Using SYCL
Performance Portability
3:53pm - 4:15pm CSTECP SOLLVE: Validation and Verification Testsuite Status Update and Compiler Insights for OpenMP
Performance Portability
4:15pm - 5:00pm CSTBlending Accelerated Programming Models in the Face of Increasing Hardware Diversity
Moderator
Back To Top Button