An optimized treatment for algorithmic differentiation of an important glaciological fixed-point problem - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Journal Articles Geoscientific Model Development Year : 2016

An optimized treatment for algorithmic differentiation of an important glaciological fixed-point problem

Abstract

We apply an optimized method to the adjoint generation of a time-evolving land ice model through algorithmic differentiation (AD). The optimization involves a special treatment of the fixed-point iteration required to solve the nonlinear stress balance, which differs from a straightforward application of AD software, and leads to smaller memory requirements and in some cases shorter computation times of the adjoint. The optimization is done via implementation of the algorithm of 5 Christianson [1994] for reverse accumulation of fixed-point problems, with the AD tool OpenAD. For test problems, the optimized adjoint is shown to have far lower memory requirements, potentially enabling larger problem sizes on memory-limited machines. In the case of the land ice model, implementation of the algorithm allows further optimization by having the adjoint model solve a sequence of linear systems with identical (as opposed to varying) matrices, greatly improving per-10 formance. The methods introduced here will be of value to other efforts applying AD tools to ice models, particularly ones which solve a " hybrid " shallow ice / shallow shelf approximation to the Stokes equations.
Fichier principal
Vignette du fichier
fp_adjoint_ice_revision1.pdf (563.46 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-01413295 , version 1 (09-12-2016)

Identifiers

  • HAL Id : hal-01413295 , version 1

Cite

Daniel Goldberg, Sri Hari Krishna Narayanan, Laurent Hascoet, Jean Utke. An optimized treatment for algorithmic differentiation of an important glaciological fixed-point problem. Geoscientific Model Development, 2016, 9 (5), pp.27. ⟨hal-01413295⟩
101 View
114 Download

Share

Gmail Facebook X LinkedIn More