You are here: Home / Groups / Short-Term Crustal Dynamics / Wiki / Work Plans / 2011-2016 Work Plans / 2011-2016 PyLith Dev Plans
3.140.197.140
  • Discoverability Visible
  • Join Policy Invite Only
  • Created 05 Jan 2021

Work Plans / 2011-2016 Work Plans /

2011-2016 PyLith Dev Plans

PyLith Development Plans

Software development plans for PyLith

Version 1.6 (by June training session)

OPTIMIZATION

  • Multiple fields in a single Sieve section done
    At this point, only used for parameters of fault constitutive models.

GENERAL

  • HDF5 output done
    Output into a single HDF5 file or an HDF5 file with raw binary files for the datasets. Datasets are written using MPI I/O if available in both cases. Add !.Xdmf metadata files associated with HDF5 files to permit reading of HDF5 files with ParaView and Visit.
  • Uniform global refinement done (98%)

QUASI-STATIC

  • Added 2-D plane strain versions of the 3-D generalized Maxwell viscoelastic rheology. done

DYNAMIC

  • Numerical damping via viscosity done
    Added numerical damping via a viscosity parameter that is independent of the bulk constitutive model. The viscosity is applied during the time integration.

COMPUTATIONAL SCIENCE

  • None

Version 1.7 (Fall 2011)

OPTIMIZATION

  • Multiple fields in a single Sieve section intermediate (0%)
    • Parameters for bulk constitutive models
    • Parameters and fields related to fault implementation
Will reduce memory use and provide speedup due to better use of memory cache.

CLEANUP

  • Separate elastic solution into elasticPrestep() function easy (0%)
  • Add —initialize_only property to Problem easy (0%)
    Permit users to stop simulation before time stepping in order to diagnose parameter settings.

GENERAL

  • Refactor initial fault tractions easy (5%)
    Create Nucleation object with spatial and temporal perturbation of tractions from initial value

QUASI-STATIC

  • Add strain hardening/softening to Drucker-Prager elastoplastic models.

DYNAMIC

  • Compute stable time step for explicit time integration easy (5%)
  • Attenuation via generalized Maxwell model (bulk and shear relaxation) intermediate (50%)

COMPUTATIONAL SCIENCE

  • Scalable mesh distribution intermediate (0%)
    Permit running large problems where the mesh cannot fit on a single compute node.
  • Accelerate FE integrations using GPUs difficult (25%)
    Will provide significant speedup to simulations run by many users because most are running on desktop machines that have GPUs.

Version 2.0

OPTIMIZATION

  • TBD

CLEANUP

  • TBD

GENERAL

  • Moment tensor point sources
    Moment tensor point sources provide a mesh independent deformation source that is better suited for Green’s function calculations than slip on a fault surface via cohesive cells. difficult

QUASI-STATIC

  • Time-step based on strain rate intermediate
  • Pressure field for incompressible elasticity problems expert

DYNAMIC

  • TBD

COMPUTATIONAL SCIENCE

  • Use KD tree search algorithm to allow output of time histories at an arbitrary location difficult
  • Use interpolated meshes (cells, faces, edges, vertices) to permit higher order basis functions expert
  • Support for arbitrary number of solution fields (e.g., displacement, temperature, pressure) expert
  • Adjust integrations to support multi-physics calculations.

Beyond Version 2.0

OPTIMIZATION

  • TBD

CLEANUP

  • TBD

GENERAL

  • Combined prescribed slip / spontaneous rupture fault condition
    Use fault constitutive model to control slip on fault except during episodes of prescribed slip. Need some way to describe when to turn on/off prescribed slip.

QUASI-STATIC

  • TBD

DYNAMIC

  • TBD

COMPUTATIONAL SCIENCE

  • TBD

Created on , Last modified on