Parthenon: A Flexible Framework for Rapid Development of Performance Portable Multiphysics Codes
MICDE University of Michigan via YouTube
Overview
Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Watch a technical seminar where Josh Dolence from Los Alamos National Laboratory presents Parthenon, an open-source framework designed for developing high-performance multiphysics simulation codes. Learn about the challenges in computational science regarding the development of state-of-the-art capabilities across diverse models, methods, and computing platforms. Explore how Parthenon addresses fundamental challenges in fields like high energy density physics and astrophysics by enabling adaptive meshes, particle simulations, and various numerical methods that can run efficiently on both CPUs and GPUs. Discover the framework's core design principles, key features, and its successful implementation across approximately 10 downstream codes. Gain insights from Dolence's extensive background in computational astrophysics and his current leadership role in the Methods for Multiscale, Multiphysics Accelerated Prediction project at LANL's Advanced Simulation and Computing Program.
Syllabus
MICDE seminar Josh Dolence: Parthenon: a flexible framework for rapid development of performance ...
Taught by
MICDE University of Michigan