.svg)
.svg)
.svg)
.svg)
.svg)
.svg)
.png)
Paper
A Portable and Efficient Lagrangian Particle Capability for Idealized Atmospheric Phenomena
Presenter
John Dennis received a PhD in computer science in 2005. He is a Scientist in the Computer Information and Systems Laboratory at the National Center for Atmospheric Research. He co-leads a research group that focuses on improving the ability of large-scale geoscience applications to utilize current and future computing platforms. His research interests include parallel algorithm and compiler optimization, graph partitioning, and data-intensive computing.
Description
The Cloud Model version 1 is an atmospheric model that allows for idealized studies of atmospheric phenomena. A new Lagrangian microphysics capability has been added, enabling a significantly more accurate representation than the traditional bulk or multi-moment approaches frequently found in mesoscale atmospheric models. We have utilized a directive-based approach to enable a single source code to efficiently support execution on both CPU and GPU-based computing platforms. In addition to the use of accelerator directives, changes to the data structures and the message-passing approach used by the Lagrangian particle-based microphysics module were necessary to enable efficient execution for a large number of particles. We focus on a configuration that will be used to investigate the impact of oceanic sea-spray on the atmospheric boundary layer within a hurricane. We observe a factor of $5.1 \times$ reduction in time to the solution when comparing the execution time for 256 NVIDIA A100 GPUs versus 256 AMD Epyc\textsuperscript{TM} Milan-based compute nodes using 1 billion particles.