Achieving High Performance with Unified Residual Evaluation

2013/09/04 by Knepley, Matthew G., Brown, Jed, Rupp, Karl +1 · 1 citation
#Computational Engineering #FOS: Computer and information sciences #Finance #Mathematical Software (cs.MS) #and Science (cs.CE)

paper · doi:10.48550/arxiv.1309.1204

Abstract

We examine residual evaluation, perhaps the most basic operation in numerical simulation. By raising the level of abstraction in this operation, we can eliminate specialized code, enable optimization, and greatly increase the extensibility of existing code.

Cited by

• Matrix-Free Evaluation Strategies for Continuous and Discontinuous Galerkin Discretizations on Unstructured Tetrahedral Grids

Related

• Negative Improvements, Relative Validity and Elusive Goodness
• Verificarlo CI: continuous integration for numerical optimization and debugging
• Numerical optimization
• Design of Computer Experiments for Optimization, Estimation of Function Contours, and Related Objectives
• Progress in Benchmarking Generics for Mathematical Computation