Software
Adaptive Mesh Refinement
Several of our adaptive mesh refinement (AMR) codes
are freely available for distribution. Others have been licensed for commercial use.
- AMRsoftware:
Mesh refinement software for 2-D and 3-D adaptive mesh refinement
for hyperbolic conservation laws.
- AMRCLAW:
An adaptive version of R. LeVeque's CLAWPACK code.
- GEOCLAW:
Formerly called TsunamiCLAW, an extension of AMRCLAW for geophysical applications
including bathymetry and other special features.
- Cart3D:
For government use, contact Michael Aftosmis (NASA Ames)
for more information. For commercial use, Cart3D has been
licensed by Desktop Aero - Aerion Corporation.
Fast Transforms and Related Codes
- NUFFT:
Fast Fourier transforms for nonequispaced data in 1,2 and 3 dimensions.
- FMPS:
Fast solvers for multiparticle electromagnetic scattering in 3 dimensions.
- Singular Quadrature Library
:
Fortran routines for integrating
singular harmonic and elastostatic kernels on triangles in 3 dimensions.
- Smooth Quadrature Library
:
Fortran routines for integrating
smooth functions on squares, triangles, cubes, and tetrahedra.
- FMM2D Libraries:
Fast multipole methods for the Laplace and Helmholtz equations in two dimensions.
- FMM3D Libraries:
Fast multipole methods for the Laplace, Helmholtz and Stokes equations in three dimensions.
- Coming Soon:
Fast Gauss transforms for summation/integration of Gaussian kernels
(statistics, machine learning, heat flow).
Computational Stochastics
We are making available some simple programs that are useful in Monte Carlo
computation.
- Acor:
A simple routine for computing error bars and autocorrelation times for time series
that come from Markov chain Monte Carlo .
- emcee - the MCMC Hammer:
Python implementation of affine-invariant stretch-move ensemble Monte Carlo sampler
Multi-core Computing Tools
A variety of parallel computing tools have been maintained or developed
in part through the CMCL.
- PyCUDA:
PyCUDA lets you access NVidia's CUDA parallel computation API from Python.
- PyOpenCL:
PyOpenCL lets you access the OpenCL parallel computation API from Python.
- MeshPy:
MeshPy offers quality triangular and tetrahedral mesh generation for Python.
- PyVisfile:
Pyvisfile allows you to write a variety of visualization file formats.
- puDB:
PuDB is a full-screen, console-based visual debugger for Python.