Year: 2016
Type: article
Abstract: The use of composable abstractions allows the application of new and established algorithms to a wide range of problems, while automatically inheriting the benefits of well-known performance optimizations. This work highlights the composition of the PETSc DMPlex domain topology abstraction with the Firedrake automated finite element system to create a PDE solving environment that combines expressiveness, flexibility, and high performance. We describe how Firedrake utilizes DMPlex to provide the indirection maps required for finite element assembly, while supporting various mesh input formats and runtime domain decomposition. In particular, we describe how DMPlex and its accompanying data structures allow the generic creation of user-defined discretizations, while utilizing data layout optimizations that improve cache coherency and ensure overlapped communication during assembly computation.
DOI: 10.1137/15m1026092
Authors: Michael Lange, Lawrence Mitchell, Matthew G. Knepley, Gerard J. Gorman
Institutions: Imperial College London, Rice University
Language: English
Subfield: Computational Mechanics
Field: Engineering
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $2,879
Awards: G102163711, Platform: Underpinning Technologies for Finite Element Simulation, G3014361174, Driving UK HPC enabled science and innovation through US collaborations 2014, SI2-SSE Collaborative Research: SPIKE-An Implementation of a Recursive Divide-and-Conquer Parallel Strategy for Solving Large Systems of Linear Equations, Gung Ho Phase 2, G7554227835, G7706340909, G8273334653
OpenAlex ID: W1153030042