forked from geodynamics/pylith
-
Notifications
You must be signed in to change notification settings - Fork 2
/
CHANGES
1481 lines (1073 loc) · 58.5 KB
/
CHANGES
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
See https://github.com/geodynamics/pylith/commits/master
for the complete log of changes made to PyLith.
======================================================================
MIGRATING FROM VERSION 2.1 TO 2.2
======================================================================
No changes are needed to parameter or other input files. The default
format for pylithinfo output is now JSON.
----------------------------------------------------------------------
Version 2.2.1
----------------------------------------------------------------------
* Added new examples.
- examples/3d/subduction: New suite of examples for a 3-D
subduction zone. This intermediate level suite of examples
illustrates a wide range of PyLith features for quasi-static simulations.
- examples/2d/subduction: Added quasi-static spontaneous rupture
earthquake cycle examples (Steps 5 and 6) for slip-weakening and
rate- and state-friction.
- These new examples make use of ParaView Python scripts to
facilitate using ParaView with PyLith.
* Improved the PyLith manual
- Added diagram to guide users on which installation method best
meets their needs.
- Added instructions for how to use the Windows Subsystem for
Linux to install the PyLith Linux binary on systems running
Windows 10.
* Fixed bug in generating Xdmf files for 2-D vector output. Converted
Xdmf generator from C++ to Python for more robust generation of Xdmf
files from Python scripts.
* Updated spatialdata to v1.9.10. Improved error messages when reading
SimpleDB and SimpleGridDB files.
* Updated PyLith parameter viewer to v1.1.0. Application and
documentation are now available on line at
https://geodynamics.github.io/pylith_parameters. Small fix to insure
hierarchy path listed matches the one for PyLith.
* Updated PETSc to v3.7.6. See the PETSc documentation for a summary
of all of the changes.
* Switched to using CentOS 6.9 for Linux binary builds to insure
compatibility with glibc 2.12 and later.
----------------------------------------------------------------------
Version 2.2.0
----------------------------------------------------------------------
* Added a browser-based parameter viewer for interactive viewing of
all PyLith parameters and version information. See Section 4.10
PyLith Parameter Viewer of the PyLith user manual.
* Adjusted packaging of the binary distributions so that they can be
used to extend PyLith and/or integrate other code with
PyLith.
* Converted the user manual from Lyx to LaTeX and added syntax
highlighting of parameter and spatial database files. Fixed several
typos.
* Fixed bug that sometimes resulted in an inconsistent fault
orientation when running in parallel. The bug appears to have been
introduced in v2.0.
* Fixed two bugs in output of solution at points that sometimes
happened in parallel simulations. The errors include:
- The order of the station names does not match the order of the
points. The point data is written in parallel by process order,
so the points for process 0 are written first, then those for
process 1, etc. This often results in reordering of the
points. The station names were written in the original order.
- The output values for some points are incorrect. The wrong cells
were being used in the interpolation.
* Updated PETSc to v3.7.5.
======================================================================
MIGRATING FROM VERSION 2.0 TO 2.1
======================================================================
The points file for OutputSolnPoints must now contain station names as
the first column.
----------------------------------------------------------------------
Version 2.1.4
----------------------------------------------------------------------
* Added --version command line argument to display version information
for PyLith and its dependencies.
* Improved information displayed with the --help command line
argument.
* Added --include-citations command line argument to display
publications to cite when publishing results from computations using
PyLith. General PyLith references are also displayed with the
--version command line argument.
* Allow use of NetCDF versions greater than 4.1.3. Switch from using
C++ API to C API.
* Fixed bug in Pythia associated with validation of parameters being
done before help could be displayed.
* Fixed typos in manual for gravity and point forces.
* Added integration with Travis for automated testing.
----------------------------------------------------------------------
Version 2.1.3
----------------------------------------------------------------------
* Add generate_statedb.py and postseismic.pvsm files missing from
examples/2d/gravity.
* Update handling of fault intersection when creating boundary
condition nodesets in examples/meshing/surface_nurbs/subduction.
* Fixes to Darwin binary package.
+ Fix linking of netCDF4 Python module.
+ Fix linking and executable mode permissions for Python scripts in
binary by using nemesis so relative links are valid.
----------------------------------------------------------------------
Version 2.1.2
----------------------------------------------------------------------
* Bugfixes for finite-strain formulation.
+ Added output of the Cauchy stresses (cauchy_stress). The second
Piola-Kirchoff stresses are output via the stress field.
+ Material properties and state variables were not retrieved
properly when updating state variables.
* Bugfixes for setting initial stress and state variables for
viscoelastic materials. The deviatoric stress state is carried
forward using the state variables, so the initial deviatoric stress
should not be considered when computing the stresses.
* Created new examples showing how to use gravity, initial stress, and
finite-strain in 2-D (examples/2d/gravity).
* Reintroduced check (that had been inadvertently removed in v2.x) for
ambiguous description of fault surface based on groups of vertices
defining faces of cells.
* Flush the output of the progress monitor so progress reports are
updated promptly.
* Updates to the user manual.
+ Added section on the debugging examples covered in recent
tutorials.
+ Added tables describing the spatial database values for each
material.
+ Included a more complete discussion of the finite-strain
formulation.
* PETSc
+ Updated to PETSc v3.7.2 (knepley/pylith branch).
+ Fixed Trilinos/ML configuration and code so that it can be built
without a Fortran compiler.
----------------------------------------------------------------------
Version 2.1.0
----------------------------------------------------------------------
* Station names are required for output at arbitrary points
(OutputSolnPoints) and are included in a /stations dataset in HDF5
files.
* A progress monitor will update a text file with the progress of a
simulation (time in the time stepping loop or the number of impulses
completed) and given an estimate of when the simulation will be
completed.
* Bug fixes
- A few bugs related to creating cohesive cells for fault
intersections have been fixed. Faults can now meet at T
intersections provided the buried edges of the faults are
clamped. In other words, the fault ending at the T intersection
has a clamped edge along the intersection. The fault ending at the
intersection must also come AFTER the through-going fault in the
list of fault interfaces.
- There have been two major bug fixes for Drucker-Prager plasticity,
for both DruckerPrager3D and DruckerPragerPlaneStrain. The first fix
was a missing initial pressure term for the plastic multiplier
in the Drucker-Prager formulation. This affects plasticity
calculations when initial stresses are used. The error has been
corrected in the code, the manual, and the unit tests. The second
bug was an incorrect test for tensile yield that could cause
PyLith to exit with an error when plastic yield had not actually
occurred. The error would only occur when the allow_tensile_yield
flag was set to False. This bug has been fixed in the code, and the
new test is also described in the manual. This should prevent
problems that previously existed when allow_tensile_yield was set
to False (as it should be for most quasi-static problems).
- Fixed bug in DataWriterHDF5Ext associated with multiple processes
writing information to the HDF5 file. With external datasets the
HDF5 file is limited to metadata and is maintained by process 0.
- A two-dimensional gravity example has been added, based on the
tutorial from the June, 2014 workshop at Stanford University.
The tutorial itself is in examples/2d/gravity, and a new section
has also been added to the manual describing the example.
- Fixed inconsistent fault orientation when running in parallel
for 2-D domains.
======================================================================
MIGRATING FROM VERSION 1.9 TO 2.0
======================================================================
Changes to various C++ objects permitted simplifying the specification
of a number of components. The map below indicates the name changes.
CellFilterAvgMesh -> CellFilterAvg
CellFilterAvgSubMesh -> CellFilterAvg
DataWriterVTKMesh -> DataWriterVTK
DataWriterVTKSubMesh -> DataWriterVTK
DataWriterVTKSubSubMesh -> DataWriterVTK
DataWriterHDF5Mesh -> DataWriterHDF5
DataWriterHDF5SubMesh -> DataWriterHDF5
DataWriterHDF5SubSubMesh -> DataWriterHDF5
DataWriterHDF5ExtMesh -> DataWriterHDF5Ext
DataWriterHDF5ExtSubMesh -> DataWriterHDF5Ext
DataWriterHDF5ExtSubSubMesh -> DataWriterHDF5Ext
Running the script:
bash $PYLITH_DIR/doc/developer/update_1.9to2.0.sh
will update all .cfg files in the current directory and all
subdirectories with the new names (you will need to replace
$PYLITH_DIR with the directory containing the PyLith source code).
PyLith allows use of the Chaco and ParMetis/Metis partitioners. The
name of the ParMetis/Metis partitioner was changed from "parmetis" to
"metis".
[pylithapp.mesh_generator]
distributor.partitioner = metis
Buried edges of faults are handled differently in v2.0. A separate
nodeset/pset should be created and contain the vertices on the buried
edges of the fault. See the Section 6.4.2 of the PyLith manual for
more information.
----------------------------------------------------------------------
Version 2.0.3
----------------------------------------------------------------------
* Bug fixes
- Updated autotools files (Makefile.am, configure.ac) for
compatibility with recent versions of automake (up to and including
v1.14.1).
----------------------------------------------------------------------
Version 2.0.2
----------------------------------------------------------------------
* Bug fixes
- Fixed linking issue in Darwin binary distribution, primarily
affecting systems with OS X 10.7 and 10.8.
- Improved example journal files for CUBIT/Trelis to improve
compatibility (examples/meshing/surface_nurbs/dem).
- Updated more journal in examples so that APREPRO lines have a
leading '$' instead of a '#' to differentiate from comments.
- Added examples/debugging files from Crustal Deformation Modeling
workshop debugging tutorial.
----------------------------------------------------------------------
Version 2.0.1
----------------------------------------------------------------------
* Bug fixes
- Improved example journal files for CUBIT/Trelis to improve
compatibility. All journal files should work with CUBIT 14.1 and
Trelis 15.0.
- Created examples of IDless journal files in examples/2d/greensfns.
These files should work with all recent versions of CUBIT and
Trelis.
- Switched journal APREPRO lines to have leading '$' instead of '#'
to differentiate from comments.
----------------------------------------------------------------------
Version 2.0.0
----------------------------------------------------------------------
* Replaced C++ Sieve implementation of finite-element data structures
with C DMPlex implementation.
DMPlex provides a simpler, more efficient implementation of the
finite-element data structures that conforms to the PETSc data
management (DM) interface. This provides tighter integration with
the rest of PETSc. Additionally, this rewrite of the data structures
results in a more efficient memory layout, resulting in better
performance.
* Improved treatment of buried fault edges, so that the slip naturally
tapers to zero along the buried edges.
An additional nodeset/pset is used to designate the buried edges of
a fault. This allows the cohesive cells to be inserted up to the
edge of the fault without splitting the mesh at the fault edge. The
slip will naturally taper to zero at along the buried edges as a
result of how the cohesive cells are created.
* Switched from using Subversion to Git for version control.
The source code repository changed from a CIG maintained Subversion
repository to a Git repository at GitHub.com. The URL for the Git
repository is https://github.com/geodynamics/pylith. The installer
has been updated accordingly.
* Added ability to recursively refine a mesh.
Global uniform refinement can now be done recursively. Each
refinement reduces the vertex spacing by a factor of two. Using more
than one level of refinement should be done carefully as the mesh
quality will generally deteriorate with more levels of refinement.
* Directories for output are created as necessary.
Directories where output files are written will be created if
necessary. Previously, the directories would not be created, so that
opening the output files in a nonexistent directory would generate
an error.
* Improved error messages.
Error messages originating in PETSc will include a stack trace that
includes both PyLith and PETSc code. Previously, only the PETSc code
was included. This provides significantly more information for
debugging.
* Improved CUBIT example for mesh sizing functions.
Based on experimentation with CUBIT 14.0, 14.1, and Trelis 15.0, we
have improved the CUBIT mesh sizing examples
(examples/meshing/cubit_cellsize). We were able to simplify the
journal files and use fewer CUBIT commands. The new procedure also
eliminates some CUBIT warnings.
* Several small improvements to various sections of the manual based
on feedback and questions from users.
- Added more information about the workflow involved in using PyLith.
- Added a discussion of how to set scales for nondimensionalization.
- Added a discussion of how the stable time step is computed for the
various materials.
- Updated and expanded the discussion of using initial state
variables.
* Bug fixes
- Fixed two MPI related bugs in computing Green's functions in
parallel. The number of impulses corresponded to only those on
process 0.
- Corrected computation of fault tractions (Lagrange multipliers) on
process boundaries for prescribed slip with explicit time stepping.
- Fixed bug when reading in list of output points with just one
point.
- Adjusted autoconf Python setup macro to remove temporary
sysconfig.pyc file.
- Added check to make sure degree of freedom specified in Dirichlet
BC is consistent with spatial dimension of problem.
- Corrected two typos in the manual related to fault opening and
tractions in examples/3d/hex8/step20 and updating to the use of
cell.dimension for the quadrature scheme with tets.
- Fixed stable time step computed for power-law viscoelastic
rheology to match manual.
======================================================================
MIGRATING FROM VERSION 1.8 TO 1.9
======================================================================
No changes are needed in .cfg files to switch from v1.8.0 to
v1.9.0. Version 1.9.0 does includes some changes to the friction and
material model interfaces, so extensions do require changes. See the
templates for details.
----------------------------------------------------------------------
Version 1.9.0
----------------------------------------------------------------------
* New features
* Added Newton-Raphson algorithm for spontaneous rupture simulations
with explicit-stepping.
Enforcing the friction criterion in a spontaneous rupture
simulation with explicit time-stepping now uses a Newton-Raphson
algorithm to find the correct traction increment. This provides a
more stable numerical solution and eliminates oscillatory behavior
when using rate-state friction.
Added SCEC spontaneous rupture benchmark TPV102 to the benchmark
repository. PyLith produces results very similar to several other
finite-element codes.
* Bug fixes
- Fixed two MPI related bugs in computing Green's functions in
parallel. The number of impulses corresponded to only those on
process 0 and the output of the impulses for vertices on processor
boundaries was inconsistent.
- Corrected computation of fault tractions (Lagrange multipliers) on
process boundaries for prescribed slip with explicit time stepping.
- Fixed bug when reading in list of output points with just one
point.
- Adjusted autoconf Python setup macro to remove temporary
sysconfig.pyc file.
- Added check to make sure degree of freedom specified in Dirichlet
BC is consistent with spatial dimension of problem.
- Corrected two typos in the manual related to fault opening and
tractions in examples/3d/hex8/step20 and updating to the use of
cell.dimension for the quadrature scheme with tetrahedral cells.
======================================================================
MIGRATING FROM VERSION 1.7 TO 1.8
======================================================================
Explicit time stepping with a non-lumped Jacobian has been eliminated
and ExplicitLumped is now Explicit.
Old setting
------------------------------------------------
formulation = pylith.problems.ExplicitLumped
formulation = pylith.problems.ExplicitLumpedTri3
formulation = pylith.problems.ExplicitLumpedTet4
New setting
------------------------------------------------
formulation = pylith.problems.Explicit
formulation = pylith.problems.ExplicitTri3
formulation = pylith.problems.ExplicitTet4
----------------------------------------------------------------------
Version 1.8.0
----------------------------------------------------------------------
* New features
* Additional flexibility in PETSc nonlinear solver parameters
The default line search type for the PETSc nonlinear (SNES) solver
is a customized backtrace method included in PyLith. The user may
now select alternative line search types (basic, bt, l2, cp)
available in PETSc.
* Post-processing utility pylith_eqinfo to compute slip information.
This post-processing utility computes the moment magnitude,
seismic moment, seismic potency, and average slip at
user-specified snapshots in time from PyLith HDF5 output.
Information is given for each fault and across all faults. See
the Post-processing section in the Running PyLith
chapter of the manual for more information.
* Computation of the stable time step for explicit time-stepping.
The stable time step for explicit time-stepping is computed based
on the CFL condition and minimum edge lengths. For triangular and
tetrahedral cells we also account for a reduction in the stable
time step due to distorted cells (e.g., slivers and needles). See
the Stable time step section in the Materials chapter of the
manual for more information.
* Output the stable time step for each cell in a material.
Output cell_info_fields "stable_dt_implicit" and
"stable_dt_explicit" can be included in material output.
* Added netCDF Python module to binary distribution to provide
Python interface to NetCDF files, including Exodus-II files. This
is used in a new meshing example for setting the discretization
size using an Exodus-II vertex field. Note that this required
updating the NetCDF library.
* Bug fixes
- Fixed omission of synchronization of stable time step computation
among processors. Minimum time step among all processors rather
than local value should be used.
- Fixed density scale not being set in NondimElasticQuasistatic.
Density scale should be set based on shear modulus, length scale,
and relaxation time.
- Added warning when initial state for a fault constitutive model is
not set. If an initial state value is not given, for rate-state
friction using a default value of L / reference slip rate. Other
fault constitutive models use a default value of 0.0 for initial
state variables.
- Separated tensor components in Xdmf files to avoid confusion. The
corresponding HDF5 files remain unchanged.
- Removed explicit time-stepping formulation with non-lumped
Jacobian. This formulation was not setup properly for spontaneous
rupture models and is too computationally expensive for practical
problems. The ExplicitLumped* formulations are now simply Explicit*.
- Fixed parallel bug that resulting in inconsistent orientation of
fault slip directions. Flipping the fault orientation was not
synchronized across processors. This bug would only appear when
running in parallel with faults that change from dipping in one
direction to dipping in the opposite direction.
- Fixed bug in setting name of field in OutputSolnPoints when output
multiple fields. This bug caused the name of the first output
field to be used and output data to overwrite each other.
======================================================================
MIGRATING FROM VERSION 1.6 TO 1.7
======================================================================
Two changes are required when migrating from version 1.6 to 1.7.
(1) The FIATSimplex object now has the same parameters as the
FIAGLagrange object.
Old setting New setting
------------------------ ------------------
cell.shape = line cell.dimension = 1
cell.shape = triangle cell.dimension = 2
cell.shape = tetrahedron cell.dimension = 3
(2) Prescribed fault tractions for spontaneous earthquake rupture use
a new, more flexible implementation that follows the same
functional form for spatial and temporal variation as that used in
the Dirichlet and Neumann boundary conditions. Consequently, the
output info fields are also different and follow the naming scheme
used in the other time-dependent boundary conditions.
Old settings
--------------------------------------------------------------------
[pylithapp.timedependent.interfaces.fault]
db_initial_tractions = spatialdata.spatialdb.SimpleDB
db_initial_tractions.iohandler.filename = tractions.spatialdb
db_initial_tractions.label = Initial fault tractions
New settings
--------------------------------------------------------------------
traction_perturbation = pylith.faults.TractPerturbation
[pylithapp.timedependent.interfaces.fault.traction_perturbation]
db_initial = spatialdata.spatialdb.SimpleDB
db_initial.iohandler.filename = tractions.spatialdb
db_initial.label = Initial fault tractions
----------------------------------------------------------------------
Version 1.7.1
----------------------------------------------------------------------
* Bug fixes
- Fixed a couple of bugs in the spontaneous earthquake rupture for
quasi-static problems when running in parallel. These prevented
the nonlinear solve from converging and erroneously generated
fault-opening in a some cases.
- Minor updates to the documentation and manual. Added Green's
function examples to the manual.
----------------------------------------------------------------------
Version 1.7.0
----------------------------------------------------------------------
* New features
* User-friendly interface for Green's functions
A new problem type provides a user-friendly interface for
computing Green's functions associated with fault slip for complex
spatial variation in elastic properties. See examples/2d/greensfns
in the tutorials for examples.
* Output of solution field at user-specified locations
Added a new output manager for interpolation of the solution field
to user-specified point locations. This feature is useful for
comparison of the solution with observations and in computing
Green's functions. See examples/3d/hex8/step19 and
examples/2d/greensfns in the tutorials for examples.
* Plane strain version of Drucker-Prager elastoplastic model
Added a plane strain version of the Drucker-Prager elastoplastic
model. Additionally, the user can now select whether to use an
inscribed, intermediate, or circumscribed fit to the Mohr Coulomb
criterion.
* Spatial and temporal variation in tractions for spontaneous
earthquake rupture
Switched from a simple constant spatial variation in initial fault
tractions to the more flexible spatial and temporal variation
consistent with the Dirichlet, Neumann, and point force boundary
conditions. Also added a switch to turn on/off applying prescribed
fault tractions when the fault opens; the default behavior is to
stop applying prescribed fault tractions when the fault opens, but
turning this off allows simulation of dike intrusions via
prescribed fault tractions. See examples/3d/hex8/step20 in the
tutorials for an example of how to specify fault tractions with
the new implementation.
* Ability to use PETSc GPU solvers
Added ability to build PyLith with either double (default) or
single precision floating point values to facilitate use of
GPUs. In order to use PETSc GPU solvers, CUDA and cusp must be
installed and PETSc must be configured to use CUDA. See the PyLith
manual and PETSc documentation for details.
* User-specified start time for simulations.
Users can set the simulation start time to any desired value. This
facilitates combining simulations to model the earthquake cycle.
* Elastic prestep in quasi-static simulations is optional.
The elastic prestep in quasi-static simulations can be skipped
(the default is to include the elastic prestep). This facilitates
combining simulations to model the earthquake cycle.
* Bug fixes
- Fixed bug in the spontaneous earthquake rupture for quasi-static
problems when running in parallel.
======================================================================
MIGRATING FROM VERSION 1.5 TO 1.6
======================================================================
No changes in parameters are required. Version 1.6.1 does require
users to specify descriptive labels for spatial databases and friction
models.
----------------------------------------------------------------------
Version 1.6.3
----------------------------------------------------------------------
* Bug fixes
- Improved error messages for problems encountered during processing
of parameters. A backtrace of the object hierarchy is now included
to pinpoint in which object the error occurred.
- Added a line search to the inner friction solve in quasi-static
simulations to increase the robustness of the nonlinear
solve. Simulations using rate and state friction now converge
under a much wider range of circumstances.
- Fixed bug in updating slip state variable in slip-weakening
friction. This caused slight errors in the cumulative slip. We
also added a parameter that forces healing to occur in a single
time step. This is used to confine slip to a single time step in
quasi-static simulations. See examples/3d/hex8/step13.cfg for an
example.
- Tuned parameters in the slip-weakening friction and rate and state
friction examples (step13.cfg and step14.cfg, respectively) in
examples/3d/hex8 to give stick-slip behavior.
- Fixed communication issue associated with writing boundary
condition information output in parallel.
- Changed info in Xdmf file for fields that are not scalars,
vectors, or tensors so that the each component is extracted,
facilitating visualization in ParaView. The corresponding HDF5
file remains the same.
- Added the ability to specify non-derived units (e.g., degree and
radian). This is useful in specifying parameters for the
Drucker-Prager elastoplastic rheologies. If no units are
specified, radians are assumed.
* Internal changes
- Rate and state friction with ageing law
The implementation of rate and state friction with ageing law was
modified to work better with the iterative solver. We switched to
the conventional, unregularized formulation but added a minimum
cutoff for the slip rate. Below this cutoff friction has a linear
rather than logarithmic dependence on slip rate. As long as this
cutoff is close to the SNES solver tolerance, the difference in
behavior is negligible while improving the ability of the solver
to converge for very small deformations.
KNOWN ISSUES
The rate and state friction with ageing law has not been tested for
dynamic rupture simulations. We plan to run the SCEC Dynamic Rupture
benchmarks for rate and state friction as soon as we add a
spatial-temporal specification of initial fault tractions, which are
required for the benchmark problems.
Running simulations with more than a million cells and large faults
in parallel can result in severe memory imbalances among
processors. Some processors around the fault may use 10x more memory
than processors away from the fault. We expect this problem to
disappear in v1.7 when we switch to new, more efficient Sieve
implementation.
----------------------------------------------------------------------
Version 1.6.2
----------------------------------------------------------------------
* Bug fixes
- Fixed bug in writing tensor data for Xdmf files. Switched Tensor
to Tensor6 to account for symmetry.
- Fixed bug in writing HDF5 files in parallel when one processor
does not write any information (e.g., faults and boundary
conditions).
- Added dimensioning of time dataset in HDF5 files. The units are
now seconds rather than nondimensional time.
- Fixed memory allocation error (std::bad_alloc) when a processor
did not contain cells for a boundary condition or output. This bug
did not show up on all architectures.
- Increased robustness of spontaneous rupture (fault friction)
implementation to broaden the range of conditions it can
handle. The implementation now properly handles cases with fault
opening and cases with zero shear or normal tractions.
* Internal changes
- Fault implementation
Several changes have been made to the fault implementation, but
none of these affect the user interface. The runtime performance
is nearly identical with improved accuracy for spontaneous rupture
(fault friction) simulations. These changes involved switching to
using tractions (non-integrated quantities) for the Lagrange
multipliers in the global coordinate system rather than integrated
quantities in the fault coordinate system. Additionally, initial
fault tractions are associated with the fault vertices and their
interpolation uses the finite-element basis functions.
- Distribution of mesh among processors
The data structures used to distribute the mesh among processors
have been improved. This reduces memory use and runtime for this
stage of the simulations.
KNOWN ISSUES
The custom line search used with the PETSc nonlinear solver (SNES)
has difficulty handling some loading cases. In cases where the
direction of the line search tends to be nearly orthogonal to the
residual, the rate of convergence in the SNES iterations is
extremely slow. In other cases the nonlinear solver gets stuck in a
local minimum. We plan to improve the line search algorithm in a
future release in order to resolve this issue and improve the rate
of convergence in spontaneous rupture simulations.
----------------------------------------------------------------------
Version 1.6.1
----------------------------------------------------------------------
* Validation of user input
Added stricter requirements for descriptive labels of various
objects, including spatial databases and friction models. The
default labels are empty strings which do not result in useful error
messages; the user is now required to specify a non-empty string for
the labels. This makes errors related to spatial databases much
easier to diagnose.
* Updates to manual
- Updated description of cell_info_fields for Neumann boundary
condition. The description had not been updated to reflect the
time-dependence introduced in version 1.4.
- Added steps 18 and 19 that discuss time-dependent Neumann boundary
conditions to examples/3d/hex8.
* Bug fixes
- Fixed bug in writing rupture information to VTK and HDF5 files
when using multiple earthquake sources. Field names did not include
name of rupture. This caused loss of information in VTK output and
a corrupted Xdmf metadata file for HDF5 output.
- Fixed error in use of initial stress tensor with generalized Maxwell
models. The initial stress tensor was added to the current stress
tensor twice.
- Fixed two bugs in the fault friction implementation. One bug
pertained to accounting for roundoff errors and convergence
tolerances in computing the slip rate. Slip rates less than
1.0e-12 (nondimensionalized) are set to zero. The friction
implementation for quasi-static problems contained a bug that
resulted in slip extending over all of the fault rather than the
appropriate isolated patch.
- Cleaned up Green's function example (examples/greensfns/hex8) so
that it runs without errors. Eliminated extraneous processing.
- Cleaned up meshing examples (examples/meshing), including
elimination of superfluous pre-processing.
- Adjusted absolute tolerances for PETSc solves in examples/3d/hex8
so that solver terminates with desired convergence criterion.
- Updated examples/2d/subduction/geometry.jou to use APREPRO
functions and variables to store id values.
----------------------------------------------------------------------
Version 1.6.0
----------------------------------------------------------------------
* New features
* Parallel binary output via HDF5
Provides much faster output by writing HDF5 files in parallel,
which can be accessed directly from Matlab or indirectly from
ParaView or Visit via automatically created Xdmf files. Temporal
data is stored in 3-D arrays, permitting slicing in time and/or
space. See examples/3d/hex8 Steps 6-9 and examples/2d/subduction
in the tutorials for examples.
* 2-D generalized Maxwell viscoelastic bulk rheology
Added a 2-D generalized Maxwell viscoelastic bulk rheology
corresponding to the plane strain version of the 3-D generalized
Maxwell viscoelastic model.
* Time-weakening fault constitutive model
Added a linear time-weakening fault constitutive model. Some
spontaneous rupture modelers prefer this model over linear
slip-weakening because it is easier to maintain resolution of the
cohesive zone.
* Global uniform parallel mesh refinement
Permits running larger problems through uniform global refinement
of the mesh by a factor of 2 (reduces the node spacing by a factor
of 2) after the mesh is distributed among processors. This allows
running problems that are 4x larger in 2-D and 8x larger in
3-D. See examples/3d/tet4 Steps 2 and 4 for examples.
* Custom algebraic multigrid preconditioner
Adds a custom preconditioner for Lagrange multiplier degrees of
freedom associated with fault slip via prescribed slip or
spontaneous ruptures with algebraic multigrid preconditioning for
quasi-static solutions. In most cases, this results in fewer
iterations in the linear solve and the number of iterations
increases much less with problem size. See examples/3d/tet4 Steps
2 and 4 for examples.
* PyLith installer utility
This utility provides a much more robust method for building
PyLith and all of its dependencies from source, including
dependency checking, installation to a central location, and
creation of a shell script to set environment variables.
* Bug fixes
- Fixed the fault friction implementation to correctly update Lagrange
multiplier values when the slip is overestimated in an
iteration. This primary fixes problems encountered with the use
of the Dieterich-Ruina rate and state fault constitutive model.
- Corrected viscoelastic rheologies to properly account for a
nonzero initial strain tensor.
======================================================================
MIGRATING FROM VERSION 1.4 TO 1.5
======================================================================
Three changes to the code require updating old parameters settings for use
with version 1.5.
(1) Recent releases of CUBIT include nodeset names in the Exodus file
and PyLith now uses them to associate vertices with boundary
conditions and faults. Use the NetCDF utility ncdump to examine the
contents of the Exodus (.exo) file to see it it includes the variable
ns_names. If it does, then use nodeset names rather than nodeset ids
for boundary condition label properties. If your Exodus file does not
contain nodeset names, then set the MeshIOCubit property
use_nodeset_names to False to continue to use nodeset id values for
boundary condition labels.
(2) The power-law constitutive parameters have been changed so that the
parameter units are no longer dependent on the power-law exponent. This
is a more logical implementation and allows (among other things) users to
vary power-law parameters using a spatial database. Previously, it was not
possible to vary power-law parameters unless everything used the same
power-law exponent. The new implementation uses reference-strain-rate,
reference-stress, and power-law-exponent to describe the material. This is
described in the 'Material Models' section of the manual.
(3) The fault property 'normal_dir' is obsolete. Only the property
'up_dir' is required to enforce that positive slip is left-lateral,
reverse, and fault-opening for dipping faults in 2-D and horizontal
fault surfaces in 3-D. Previously, in 2-D positive slip was always
left-lateral, but now the up-direction is used to enforce positive
slip corresponds to reverse motion for dipping faults. For horizontal
fault surfaces in 3-D a normal of (0,0,1) is assumed in determining
the up-dip direction.
----------------------------------------------------------------------
Version 1.5.2
----------------------------------------------------------------------
* PyLith 1.5.2 requires FIAT version 0.9.9 or later and an updated
PETSc development version. It also requires users to update to the
latest spatialdata version for compatibility of the SWIG generated
files. These are included in the binary distribution, but users
building PyLith from source will need to update FIAT, PETSc, and
spatialdata.
* Bug fixes
- Fixed setting of elastic constants in DruckerPrager3D and
computation of the yield function. Some off-diagonal elasticity
constants were off by a factor of 2.0 and the yield function was
missing a factor of 0.5 and sqrt().
- Fixed computation of stable time step when using initial
stresses with PowerLaw3D. If effective stress is zero, then
stable time step is infinite.
- Re-enabled check for compatibility of quadrature scheme and
cells for bulk rheologies.
- Added check to configure for compatible version of FIAT.