Wednesday, 16:15 - 16:40 h, Room: H 1058


Eric Phipps
Support embedded algorithms through template-based generic programming

Coauthors: Roger Pawlowski, Andy Salinger


We describe a framework for incorporating embedded analysis algorithms, such as derivative-based optimization and uncertainty quantification, in large-scale simulation codes using template-based generic programming. The framework is based on standard C++ language constructs such as templating, operator overloading, expression templates, and template metaprogramming, and enables the incorporation of advanced algorithms with a minimum of programmer effort. In this talk we describe the overall approach, several software tools implementing the approach in the Trilinos solver framework, and examples demonstrating the usefulness of the approach applied to optimization and uncertainty quantification of large-scale PDE-based simulations.


Talk 3 of the invited session Wed.3.H 1058
"Software for large-scale optimization" [...]
Cluster 10
"Implementations & software" [...]


  To apply for USA Payday Loans you don't have to seek the help of relatives or go to a bank. If you have already decided to take Generic Levitra, be sure to consult a doctor, you don't have any contraindications and act strictly due to a prescription.