[cig-commits] r17039 - in short/3D/PyLith/trunk: . examples/3d/hex8

[brad at geodynamics.org]

Author: brad
Date: 2010-07-13 10:08:53 -0700 (Tue, 13 Jul 2010)
New Revision: 17039

Modified:
   short/3D/PyLith/trunk/TODO
   short/3D/PyLith/trunk/examples/3d/hex8/step02.cfg
Log:
Updated TODO.

Modified: short/3D/PyLith/trunk/TODO
===================================================================
--- short/3D/PyLith/trunk/TODO	2010-07-12 22:45:22 UTC (rev 17038)
+++ short/3D/PyLith/trunk/TODO	2010-07-13 17:08:53 UTC (rev 17039)
@@ -2,18 +2,19 @@
 CURRENT ISSUES/PRIORITIES
 ======================================================================
 
-----------------------------------------------------------------------
-SECONDARY PRIORITIES
-----------------------------------------------------------------------
+* Uniform global refinement for tets with faults
 
-* Better preconditioning [Matt/Brad]
-
-  Need field split working for both SolverLinear and SolverNonlinear.
-
 * Output to HDF5 files.
 
-* Uniform global refinement for tets with faults
+* Reformulate solves in terms of Schur complement
 
+  + adjustSolnLumped()
+    Compute Lagrange multipliers then displacements
+  + form C as sparse matrix, use PETSc operations for computation
+    (will work parallel)
+  + Refactor friction sensitivity solve, fault preconditioner, and
+    adjustSolnLumped()
+
 * Cleanup
   + memory model
   + full-scale testing
@@ -31,10 +32,23 @@
   + Specialized elasticity integrator objects for Quad4, Hex8
     - Add for lgdeform, implicit time integration?
 
-* Testing
-  + Need full-scale test with variation in physical properties
-    within a material.
+* Lumped elasticity formulation for large deformations
 
+----------------------------------------------------------------------
+SECONDARY PRIORITIES
+----------------------------------------------------------------------
+
+* Combined kinematic and dynamic fault implementation
+
+  FaultCohesiveKinDyn
+
+  Kinematic slip in confined space/time region, otherwise governed by
+  fault constitutive model.
+
+* Better preconditioning [Matt/Brad]
+
+  Need field split working for both SolverLinear and SolverNonlinear.
+
 * utilities
   + pylith_eqsummary
     Calculate eq statistics for given slip model (VTK output).
@@ -42,18 +56,6 @@
 
 * GenMaxwellPlaneStrain (plane strain Generalize Maxwell model) [Charles]
 
-* VTKDataReaderZip (obsolete if use VTK XML files?)
-
-  Read VTK files from zipfile.
-
-  import zipfile
-  zip = zipfile.open("test.zip")
-  for filename in zip.namelist():
-    bytes = zip.read(filename)
-    reader.input_string = bytes
-    reader.update()
-    data = reader.get_output()
-
 * Lumped solver [Brad]
   + Need to finish unit tests
 
@@ -65,8 +67,6 @@
   Make fault nucleation (initial tractions) modular (allow space/time
   variation).
 
-* Lumped elasticity formulation for large deformations
-
 * Manual
 
   Are material models consistent with governing equations discussion?
@@ -112,9 +112,6 @@
       problems C++, SWIG
       topology C++, SWIG
 
-1. Savage-Presscott benchmark
-    Tet mesh
-
 2. Analytic BC for reverse-slip benchmark [Charles or Brad]
     a. Create BC for CUBIT mesh.
     b. Create BC for LaGriT mesh.
@@ -215,8 +212,6 @@
 RELEASE 1.5
 ----------------------------------------------------------------------
 
-2. 2-D Plane strain Maxwell viscoelastic rheology [Charles]
-
 3. 2-D Plane strain Generalized Maxwell viscoelastic rheology [Charles]
 
 4. 2-D Power-law rheology [Charles]
@@ -235,11 +230,6 @@
 7. Body forces? [Charles is going to discuss this with Brad Hager to
 see if this will satisfy their needs]
 
-Modularize output? [Matt]
-
-Uniform refinement (debug, check, all cell types)
-
-
 ======================================================================
 KNOWN DEFICIENCIES
 ======================================================================
@@ -309,12 +299,8 @@
      Interpolate solution to desired location
      Solution field (time history)
 
-  3. Output velocity field
-
   4. Uniform global refinement
 
-  5. Drucker-Prager viscoelastoplastic
-
   6. FaultCohesiveTract (dikes)
 
 ----------------------------------------------------------------------

Modified: short/3D/PyLith/trunk/examples/3d/hex8/step02.cfg
===================================================================
--- short/3D/PyLith/trunk/examples/3d/hex8/step02.cfg	2010-07-12 22:45:22 UTC (rev 17038)
+++ short/3D/PyLith/trunk/examples/3d/hex8/step02.cfg	2010-07-13 17:08:53 UTC (rev 17039)
@@ -77,6 +77,10 @@
 db_initial.label = Neumann BC on +x
 db_initial.iohandler.filename = spatialdb/tractions_axial_shear.spatialdb
 
+output.cell_info_fieldsia = [initial-value]
+output.writer.filename = output/step02-traction.vtk
+output.cell_filter = pylith.meshio.CellFilterAvgSubMesh
+
 # We must specify quadrature information for the cell faces.
 quadrature.cell = pylith.feassemble.FIATLagrange
 quadrature.cell.dimension = 2