YoVDO

High-Order Solvers and GPU Acceleration for MFEM - Workshop 2022

Offered By: Inside Livermore Lab via YouTube

Tags

Finite Element Method Courses Numerical Analysis Courses Computational Physics Courses GPU Acceleration Courses MFEM Courses

Course Description

Overview

Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Explore the benefits and challenges of high-order methods in scientific simulations through this conference talk from the MFEM Workshop 2022. Delve into Will Pazner's presentation on "High-Order Solvers + GPU Acceleration," which covers the advantages of high-order methods for modeling under-resolved physics and their performance on modern computing architectures. Learn about MFEM's support for matrix-free solvers in high-order methods, the setup and application of high-order operators, and low-order-refined (LOR) preconditioning and matrix assembly. Gain insights into LOR assembly throughput on GPUs, including comparisons between CPU and GPU performance and parallel scalability. Discover the potential of LOR adaptive mesh refinement preconditioning in enhancing simulation accuracy and efficiency.

Syllabus

MFEM Workshop 2022 | High-Order Solvers + GPU Acceleration


Taught by

Inside Livermore Lab

Related Courses

The Finite Element Method for Problems in Physics
University of Michigan via Coursera
Computational Physics
Indian Institute of Science Education and Research, Pune via Swayam
Manipulate Coulomb's Law Concepts using Wolfram notebook
Coursera Project Network via Coursera
Lagrangian Coherent Structures in Unsteady Fluids with Finite Time Lyapunov Exponents
Steve Brunton via YouTube
Changing Physics Education with Julia
The Julia Programming Language via YouTube