hyteg merge requestshttps://i10git.cs.fau.de/hyteg/hyteg/-/merge_requests2020-10-12T11:13:41+02:00https://i10git.cs.fau.de/hyteg/hyteg/-/merge_requests/370"Horbach" benchmark2020-10-12T11:13:41+02:00Marcus Mohr"Horbach" benchmarkHi,
the implementation of the "Horbach" benchmark is not fully complete, yet. There is some work on the solvers that still needs to be done. However, there are some updates to the StokesOperator, StokesBlockPreconditioner and the FreeSl...Hi,
the implementation of the "Horbach" benchmark is not fully complete, yet. There is some work on the solvers that still needs to be done. However, there are some updates to the StokesOperator, StokesBlockPreconditioner and the FreeSlipWrapper in the P2P1Elementwise setting that IMHO should already be incorporated into the master.
Cheers
MarcusMarcus MohrMarcus Mohrhttps://i10git.cs.fau.de/hyteg/hyteg/-/merge_requests/365Extends functionality of and for vector functions2020-09-28T14:49:01+02:00Marcus MohrExtends functionality of and for vector functionsHi,
after the merge one can directly, i.e. by one method/function call,
- synchronise a vector function over primitives
- interpolate its components
- add them to a VTKOutput object, which will export them as vector fields
- query thei...Hi,
after the merge one can directly, i.e. by one method/function call,
- synchronise a vector function over primitives
- interpolate its components
- add them to a VTKOutput object, which will export them as vector fields
- query their name and dimensionality
Cheers
MarcusMarcus MohrMarcus Mohrhttps://i10git.cs.fau.de/hyteg/hyteg/-/merge_requests/363DiagonalNonConstantOperator2020-09-23T13:14:29+02:00Marcus MohrDiagonalNonConstantOperatorHi,
the merge will add a new operator to the master. So far we could only generate diagonal operators, i.e. ones that only work with the diagonal of the associated matrix respectively the central stencil weight in the case of constant s...Hi,
the merge will add a new operator to the master. So far we could only generate diagonal operators, i.e. ones that only work with the diagonal of the associated matrix respectively the central stencil weight in the case of constant stencils. The new operator extends this to the case of non-constant stencils.
Cheers
MarcusNils KohlNils Kohlhttps://i10git.cs.fau.de/hyteg/hyteg/-/merge_requests/343Mixed forms for P2-P1 Stokes with blending2020-07-29T16:49:50+02:00Marcus MohrMixed forms for P2-P1 Stokes with blendingCloses #126Closes #126Marcus MohrMarcus Mohrhttps://i10git.cs.fau.de/hyteg/hyteg/-/merge_requests/342Draft: Resolve "Implement forms needed for P2-P1 Stokes with blending"2020-07-29T16:48:23+02:00Marcus MohrDraft: Resolve "Implement forms needed for P2-P1 Stokes with blending"Closes #126Closes #126https://i10git.cs.fau.de/hyteg/hyteg/-/merge_requests/325Mixed HyTeG forms2020-04-01T19:09:15+02:00Marcus MohrMixed HyTeG formsHi,
in order to run simulations for the Stokes problem with P2-P1 Taylor Hood elements in combination with blending we need forms for the gradient and divergence part. This merge brings in the corresponding functionality, which are *mix...Hi,
in order to run simulations for the Stokes problem with P2-P1 Taylor Hood elements in combination with blending we need forms for the gradient and divergence part. This merge brings in the corresponding functionality, which are *mixed forms*. Because that was straightforward I also added the forms for the P2-P2 case. Note that currently there is no support for 3D.
The new forms are intended to be used with the mixed elementwise operators Nils recently implemented for this application scenario.
Cheers
MarcusMarcus MohrMarcus Mohrhttps://i10git.cs.fau.de/hyteg/hyteg/-/merge_requests/318New HyTeG forms and GeometryMaps2020-03-09T13:38:45+01:00Marcus MohrNew HyTeG forms and GeometryMapsNils KohlNils Kohlhttps://i10git.cs.fau.de/hyteg/hyteg/-/merge_requests/306Matrix Setup for Elementwise Operators2020-02-04T10:17:36+01:00Marcus MohrMatrix Setup for Elementwise OperatorsHi,
the merge adds matrix assembly for elementwise operator via our PETSc interface.
Cheers
MarcusHi,
the merge adds matrix assembly for elementwise operator via our PETSc interface.
Cheers
MarcusMarcus MohrMarcus Mohrhttps://i10git.cs.fau.de/hyteg/hyteg/-/merge_requests/295Elementwise operators2020-01-08T16:07:03+01:00Marcus MohrElementwise operatorsHi,
this is a new attempt to have elementwise operators, i.e. operators that do not work with a stencil-based paradigm, but employ the local element matrices directly. Thus, the apply() will compute the effect of multiplying the element...Hi,
this is a new attempt to have elementwise operators, i.e. operators that do not work with a stencil-based paradigm, but employ the local element matrices directly. Thus, the apply() will compute the effect of multiplying the element local DoFs with the local element matrix. Currently by setting the latter up, but in the future could also use the technique of Kronbichler et al.
This new attempt implements a P1ElementwiseOperator and a P2ElementwiseOperator for 2D and 3D. So far we can only apply the operator and perform Jacobi smoothing.
Cheers
MarcusMarcus MohrMarcus Mohrhttps://i10git.cs.fau.de/hyteg/hyteg/-/merge_requests/285Resolves issue #1122019-11-26T15:31:40+01:00Marcus MohrResolves issue #112Hi,
this merge will close issue #112 (again :wink:). We now also have ````DGFunction.multElementwise()```` available and after rebasing the pipeline for the merge should run through.
Cheers
MarcusHi,
this merge will close issue #112 (again :wink:). We now also have ````DGFunction.multElementwise()```` available and after rebasing the pipeline for the merge should run through.
Cheers
MarcusMarcus MohrMarcus Mohrhttps://i10git.cs.fau.de/hyteg/hyteg/-/merge_requests/283Resolves issue #1122019-11-25T13:47:23+01:00Marcus MohrResolves issue #112Hi,
I implemented the EdgeDoFFunction::multElementwise() method which was missing to also have this functionality for P2Functions and other functions with vertex and edge dofs. I also added another test routine for this feature in 2D an...Hi,
I implemented the EdgeDoFFunction::multElementwise() method which was missing to also have this functionality for P2Functions and other functions with vertex and edge dofs. I also added another test routine for this feature in 2D and 3D.
Cheers
MarcusNils KohlNils Kohlhttps://i10git.cs.fau.de/hyteg/hyteg/-/merge_requests/272Petsc block preconditioning with rebase2019-11-06T14:26:32+01:00Nils KohlPetsc block preconditioning with rebasehttps://i10git.cs.fau.de/hyteg/hyteg/-/merge_requests/259Tools for Spectrum Estimation2019-09-18T12:52:35+02:00Marcus MohrTools for Spectrum EstimationHi,
on the mohr/numericTools branch I implemented two functions for computing estimates for the eigenvalue spectrum of an operator.
Cheers
MarcusHi,
on the mohr/numericTools branch I implemented two functions for computing estimates for the eigenvalue spectrum of an operator.
Cheers
MarcusNils KohlNils Kohlhttps://i10git.cs.fau.de/hyteg/hyteg/-/merge_requests/246mohr/issue992019-07-17T13:06:59+02:00Marcus Mohrmohr/issue99Follow up on merge request !240 which is superseded by this one. Will fix issue #99.Follow up on merge request !240 which is superseded by this one. Will fix issue #99.Marcus MohrMarcus Mohrhttps://i10git.cs.fau.de/hyteg/hyteg/-/merge_requests/240[WIP] mohr/integrate general forms into master2019-07-17T12:42:45+02:00Marcus Mohr[WIP] mohr/integrate general forms into masterHi,
working on issue #99. Things that need to be done:
- [x] merge general forms into master fixing merging issues
- [x] make sure compilation of framework, apps and tests works
- [x] implement parts that are missing for assembling P2 ...Hi,
working on issue #99. Things that need to be done:
- [x] merge general forms into master fixing merging issues
- [x] make sure compilation of framework, apps and tests works
- [x] implement parts that are missing for assembling P2 stencils in 3D
- [x] make sure there are no more errors in test suite
Cheers
MarcusMarcus MohrMarcus Mohrhttps://i10git.cs.fau.de/hyteg/hyteg/-/merge_requests/231Implements inline mesh generation for rectangular cuboids.2019-04-18T16:05:07+02:00Marcus MohrImplements inline mesh generation for rectangular cuboids.The commit implements a new function MeshInfo::meshCuboid() that
allows inline generation of a tetrahedral mesh for a rectangular
cuboid.
A demo using the new mesher is added to the show_mesh app.
Note that the current limitation is th...The commit implements a new function MeshInfo::meshCuboid() that
allows inline generation of a tetrahedral mesh for a rectangular
cuboid.
A demo using the new mesher is added to the show_mesh app.
Note that the current limitation is that the cuboid is axi-aligned.
However, this only affects computation of the vertex coordinates and
could easily be changed.Marcus MohrMarcus Mohrhttps://i10git.cs.fau.de/hyteg/hyteg/-/merge_requests/229P2 3D grid transfer2019-04-17T10:23:20+02:00Nils KohlP2 3D grid transferNils KohlNils Kohlhttps://i10git.cs.fau.de/hyteg/hyteg/-/merge_requests/220Implemented application of Dirichlet BCs to a linear system without losing...2019-04-02T11:10:20+02:00Nils KohlImplemented application of Dirichlet BCs to a linear system without losing...Implemented application of Dirichlet BCs to a linear system without losing symmetry for PETSc solvers.
Uses the PETSc function MatZeroRowsColumns() which does that automatically.
Still, we need to think how we can easily integrate this ...Implemented application of Dirichlet BCs to a linear system without losing symmetry for PETSc solvers.
Uses the PETSc function MatZeroRowsColumns() which does that automatically.
Still, we need to think how we can easily integrate this to use more efficient
solvers in HyTeG, because the RHS is modified depending on the original system.
Possible solutions:
1. re-assemble the system every time we solve it since we need to also rebuild the RHS
2. store a copy of the original system and circumvent re-assembling by copying it and applying
only MatZeroRowsColumns() (without re-assembly) before calling the solver -
If PETSc is only used as a coarse grid solver, option 2 might be a good solutionNils KohlNils Kohlhttps://i10git.cs.fau.de/hyteg/hyteg/-/merge_requests/219Multigrid experiments2019-04-01T11:37:07+02:00Nils KohlMultigrid experimentsNils KohlNils Kohlhttps://i10git.cs.fau.de/hyteg/hyteg/-/merge_requests/217P2 3D SOR2019-04-16T19:15:45+02:00Nils KohlP2 3D SORNils KohlNils Kohl