11

Capabilities:

• Building blocks for block-structured AMR codes for solving time-dependent PDE’s• Functionality for [1…6]D, mixed-dimension building• Supports data on cell centers, faces and nodes.• Support for

Grid-based operations (explicit and/or implicit) Communication between levels (explicit and/or implicit) Particle and particle/mesh algorithms

• Support for second- and higher-order time-stepping methods• Hybrid C++/Fortran 90 code. ChomboFortran source-to-source DSL • Can run in serial, with MPI only, or with hybrid MPI + OpenMP for node-level parallelism• Has native single-level and multilevel multigrid solvers that respect the inherent domain

decomposition• HDF5 I/O for both restart and plotfiles – visualization format supported by Visit.

Download: anag-repo.lbl.gov

Further information: Brian Van Straalen, bvstraalen@lbl.gov

Chombo – Block-structured AMR software framework

22

Large Eddy simulations of wind turbines (PI, R. Samtaney, KAUST, S. Guzik CSU).

CHARM - Compressible CFD + collisionless particle cosmology simulations (PI F. Miniati, ETH Zurich)

MS_FLUKSS - Physics of the Solar Wind and Its Interaction with the Interstellar medium using compressible hyperbolic CFD + electromagnetic and kinetic effects (G. Zank. and N. Pogorelov UA-Huntsville)

PLUTO - General astrophysics modeling (PI A. Mignone) ORION - Astrophysical MHD turbulence PI C. McKee / R. Klein, UCB REALM - SF Bay and Delta Hydrology modeling shallow water (PI P. Schwartz,

LBNL and CA Department of Water Resources) Plasma-wakefield accelerators -- compressible viscous flow (PI D. Graves,

LBNL). ChomboCrunch - Pore-scale subsurface reacting flow code (PI D. Trebotich,

LBNL) Conjugate heat transfer in nuclear reactors (PI R. Crockett, LBNL, TechX) COGENT - 4D gyrokinetic models of tokamak edge ( PI J. Hittinger, LLNL) BISICLES - Land ice model for climate simulation (PI D. Martin, LBNL)

FASTMath SciDAC Institute

Chombo: Application Codes

33

Chombo-Crunch:

CFD + multi-component geochemical reactive transport in very complex pore (micro) scale geometries Adaptive, finite volume methods for advection-

diffusion in Chombo• Accurate reactive surface area using

embedded boundaries• Dynamic local refinement (AMR)• Scalable (100K processors)• Direct simulation of image data

CrunchFlow geochemistry• SNIA operator splitting• Point-by-point calculation• CFL-limited timestep

Rate calculated at each water-mineral interface by multiplying by the reactive surface area (RSA)

Computational domain for calcite in capillary tubeImage data converted to simulation grid

using implicit function representation of boundaries