Automatic differentiation of c++ codes for large-scale scientific computing

Authors:
Roscoe A. Bartlett;David M. Gay;Eric T. Phipps
Affiliations:
Sandia National Laboratories, Albuquerque, NM;Sandia National Laboratories, Albuquerque, NM;Sandia National Laboratories, Albuquerque, NM
Venue:
ICCS'06 Proceedings of the 6th international conference on Computational Science - Volume Part IV
Year:
2006

Citing 3
Cited 2

Evaluating derivatives: principles and techniques of algorithmic differentiation

Evaluating derivatives: principles and techniques of algorithmic differentiation
Integrating AD with object-oriented toolkits for high-performance scientific computing

Automatic differentiation of algorithms
Adifor 2.0: Automatic Differentiation of Fortran 77 Programs

IEEE Computational Science & Engineering

Graph-Based Software Design for Managing Complexity and Enabling Concurrency in Multiphysics PDE Software

ACM Transactions on Mathematical Software (TOMS)
Automating embedded analysis capabilities and managing software complexity in multiphysics simulation, Part I: Template-based generic programming

Scientific Programming - A New Overview of the Trilinos Project --Part 1

Quantified Score

Hi-index	0.00

Visualization

Abstract

We discuss computing first derivatives for models based on elements, such as large-scale finite-element PDE discretizations, implemented in the C++ programming language. We use a hybrid technique of automatic differentiation (AD) and manual assembly, with local element-level derivatives computed via AD and manually summed into the global derivative. C++ templating and operator overloading work well for both forward- and reverse-mode derivative computations. We found that AD derivative computations compared favorably in time to finite differencing for a scalable finite-element discretization of a convection-diffusion problem in two dimensions.