Skip to content
Lagrangian finite-element code for solid mechanics on next-generation computing platforms.
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
cmake
docs Removed html directory Nov 8, 2018
examples
scripts Adding majority of source code and test files. Jun 25, 2018
src Removed use of depreciated exodus routines from nimble_exodus_output.cc. Apr 4, 2019
test
.gitignore Added cache directory for ccls to .gitignore Feb 10, 2019
AUTHORS Adding majority of source code and test files. Jun 25, 2018
CMakeLists.txt Move USE_PURE_MPI switch detection before bvh library support in CMak… Mar 7, 2019
README.md Expanded readme file. Jun 25, 2018

README.md

NimbleSM

NimbleSM is a Lagrangian finite-element code for solid mechanics. Its primary application is the solution of mechanics problems on nonuniform, three-dimensional meshes using either explicit transient dynamic or implicit quasi-static time integration. Additional features include multiscale modeling and the simulation of contact. The NimbleSM code base is designed for performance portability across varying hardware architectures.

NimbleSM may be built with the following optional dependencies:

**** Exodus

Library for I/O. It is highly recommended that you build NimbleSM with exodus support. The alternative is an exceedingly slow text-based I/O system.

https://github.com/gsjaardema/seacas

**** Kokkos

Performance portability programming ecosystem. Utilized by NimbleSM for improved performance on GPU-based systems and other emerging hardware platforms.

https://github.com/kokkos/kokkos

**** Qthreads

Lightweight locality-aware user-level threading runtime. Utilized by NimbleSM for improved performance via on-node accelerators.

https://github.com/Qthreads/qthreads

**** Trilinos

Algorithms for the solution of large-scale engineering and scientific problems. Utilized by NimbleSM for parallel linear solvers. Required for parallel execution of implicit time integration (quasi-statics).

https://trilinos.org

**** Darma

An asynchronous many task (AMT) programming model and runtime. Not publicly available.

https://github.com/darma-tasking

You can’t perform that action at this time.