[deal.II] Periodic boundary constraints with arbitrary transformation of field

Hi all, Cutting to the chase: my problem involves using traceless, symmetric 3x3 tensor fields, and I represent these as 5-component vector finite element fields, where each vector-component is a degree of freedom of the tensor. My domain is a cylinder, and I'm trying to set up periodic boundar

[deal.II] Distance between points across periodic boundary

Hi everyone, I have a problem whereby I must calculate whether points are in a neighborhood of some base point (e.g. within some L2 distance). In the bulk this is simple, but I'm wanting to impose periodic boundary conditions on some of the edges. In this case, it's conceivable that a base poin

Re: [deal.II] Determine location of a degree of freedom

Hey no worries, I have also been caught posting questions that were in the FAQ. - Lucas On Fri, Jul 28, 2023, 4:59 PM Sean Johnson wrote: > I'm so sorry that it was in the FAQ. > > Thanks for the fast response. I will try this immediately. Thanks again > > On Friday, July 28, 2023 at 3:56:47 PM

Re: [deal.II] Determine location of a degree of freedom

Hi Sean, Is this what you're looking for? https://github.com/dealii/dealii/wiki/Frequently-Asked-Questions#how-to-get-the-mapped-position-of-support-points-of-my-element - Lucas On Fri, Jul 28, 2023, 4:51 PM Sean Johnson wrote: > Is there a way to findout the "coordinates" of a degree of free

Re: [deal.II] Integrate difference isn't picking on the

I haven't tried this with deal.II yet, but I've recently started using the mold linker for programs I write which have many compilation units (link to GitHub here: https://github.com/rui314/mold). If you're only changing one line then probably only one compilation unit is being recompiled, so the b

Re: [deal.II] Weak scaling running out of memory

a smaller scale, > even though it might take 2-3 attempts to have the timers in the right > spots. > > Best, > Martin > On 07.06.23 21:04, Lucas Myers wrote: > > Hi everyone, > > I'm trying to run a scaling analysis on my code, but when I make the > system

[deal.II] Weak scaling running out of memory

Hi everyone, I'm trying to run a scaling analysis on my code, but when I make the system too large I get a (Killed) error, which I strongly suspect has to do with memory issues. I'm unsure why this is happening, because I am increasing resources proportional to the system size. Details: I'm t

Re: [deal.II] Error in using FESystem usage

Hi Vinayak, If you do not manually initialize member variables in a constructor, they are automatically initialized with their default constructor (i.e. one which takes no arguments). However, the FESystem class is such that the default constructor is manually deleted, and so can't be constructed

[deal.II] Internal deal.II error when instantiating TrilinosWrappers::block_operator

Hi folks, I'm trying to use the block_operator technique on Trilinos vectors and matrices, but I can't quite get it to compile. For simplicity, the relevant lines of code are: ``` #include #include using vec = dealii::LinearAlgebraTrilinos::MPI::Vector; using block_vec = dealii::LinearAlgebra

[deal.II] Preconditioners not lowering the number of GMRES iterations

Hi everyone, TL;DR: I'm trying to precondition my system which can be solved with GMRES (with increasing iteration number for increasing size) but the standard preconditioners are either increasing the number of iterations, or causing the solver not to converge. Details: I'm trying to solve a

Re: [deal.II] Integrate on local subsets of domain on distributed triangulation

Hi Wolfgang, This is extremely helpful, thanks so much! - Lucas On Tuesday, April 25, 2023 at 11:24:31 AM UTC-5 Wolfgang Bangerth wrote: > On 4/25/23 09:44, Lucas Myers wrote: > > > > Re: FEPointEvaluation and RemotePointEvaluation, I think I actually need > > more tha

Re: [deal.II] Integrate on local subsets of domain on distributed triangulation

Hi Wolfgang, Thanks for the help! Point taken about neighboring processes and communication latency. Re: FEPointEvaluation and RemotePointEvaluation, I think I actually need more than that because I need the quadrature points on each of the cells that I'm querying, so I can just do the standar

[deal.II] Re: Integrate on local subsets of domain on distributed triangulation

Ack, I realize the pasted screenshot of the relevant equation has disappeared! I've included it as an attachment to this reply. - Lucas On Saturday, April 22, 2023 at 2:49:07 PM UTC-5 Lucas Myers wrote: > Hi folks, > > I'd like to coarse-grain a finite element field, where

[deal.II] Integrate on local subsets of domain on distributed triangulation

Hi folks, I'd like to coarse-grain a finite element field, where the coarse-graining procedure is just convolving the field with a Gaussian function. More concretely, I'd like to do the following: For a regular Triangulation, we can calculate at every quadrature point by iterating through th

Re: [deal.II] Biharmonic equation with periodic BCs

Hi Wolfgang, Thanks for the feedback! I have also not found any literature dealing with this periodic boundary conditions problem -- I doubt I will try to get to the bottom of it experimentally unless I run into technical issues solving my particular problem. With both the lifting method and in

[deal.II] Biharmonic equation with periodic BCs

Hi folks, Given the explanations in step-47 and step-82 , it appears that the particular boundary conditions associated with the biharmonic equation can make it relatively easy to

[deal.II] Status of hybrid shared/distributed parallelism

Hi folks, I'm wondering if there's somewhere I can look to get a broad overview of the different parallelization schemes available in deal.II for various pieces of the library, and maybe what people's experiences have been. As I understand it, any matrix/vector assembly can be done with a shar

[deal.II] Re: Get DoFHandler::cell_iterator from Triangulation::cell_iterator

Oops, this is in the FAQ: https://github.com/dealii/dealii/wiki/Frequently-Asked-Questions#can-i-convert-triangulation-cell-iterators-to-dofhandler-cell-iterators My bad! - Lucas On Thursday, February 2, 2023 at 3:53:07 PM UTC-6 Lucas Myers wrote: > Hi everyone, > > I'm try

[deal.II] Get DoFHandler::cell_iterator from Triangulation::cell_iterator

Hi everyone, I'm trying to use the `distributed_compute_point_locations` functions to compute a finite element function's values at specific points. However, the cells that this function returns are of type Triangulation::cell_iterator, and in order to use the `FEValues::get_function_values` fu

[deal.II] Mechanisms for deleting cells/edges/vertices in some region

Hi folks, I'm trying to construct a rectangular grid with features at arbitrary locations. The most straightforward way of doing this that I can think of is to generate a rectangular hypercube, delete the cells/edges/vertices around feature locations, and then attach other triangulations (e.g.

Re: [deal.II] constraints.distribute does not seem to be enforcing Dirichlet conditions

how to continue debugging. - Lucas On Tuesday, August 30, 2022 at 3:48:06 PM UTC-5 Wolfgang Bangerth wrote: > On 8/30/22 13:56, Lucas Myers wrote: > > > > To enforce these boundary conditions, I initialize constraints with: > > ``` > > constraints.clear(); > >

[deal.II] constraints.distribute does not seem to be enforcing Dirichlet conditions

Hi everyone, I'm trying to solve a nonlinear ODE with Dirichlet conditions using Newton's method. For this I give an initial guess for the solution ( y = (1/2) x ) and then update each Newton step with an update that ought to be zero at the boundaries (hence preserving the boundary conditions o

[deal.II] Run solver independently in different subdomains

Hi everyone, I've got a diffusion-ish equation where I'd like to do the following: 1. Partition my domain into n subdomains -- for concreteness, say the domain is a rectangle and my partitions are two (completely separated) circles along with the rest of the rectangle. 2. Relax the

[deal.II] Suggestions for finding local minima on distributed, adaptively-refined meshes

Hi folks, I suspect this is a simple task, but I can't find anything which does it outright. I'd like to find the local minima associated with some quantity derived from a finite element field. The features that I'm looking at are reasonably well-behaved, in that they're smooth except exactly a

[deal.II] Deleted move constructors and inability to change objects post-initialization

Hi everyone, Is there any way to initialize a Simulation object (akin to the StepX objects found in the tutorials) in a method and then return it via std::move()? Or, barring that, is there a way to reinitialize an FESystem member variable to have a different degree? In the former case I run

Re: [deal.II] Multigrid in 3D questions

tigrid (geometric or algebraic) to the entire coupled >> problem is just fraught with so many problems that it is hard to get >> right and harder yet to efficiently implement. >> >> Best >> W. >> >> >> On 3/3/22 10:22, Lucas Myers wrote: >>

[deal.II] Multigrid in 3D questions

Hi folks, I'm trying to solve a 3D problem in parallel using an AMG preconditioner, but the performance is bad. I'm wondering if I can get some advice from someone who has experience choosing and tuning multigrid preconditioners, particularly for problems in 3D. For context: - The problem

Re: [deal.II] Re: get_function_values giving different results depending on the number of MPI ranks

Done. Issue is documented here <https://github.com/dealii/dealii/issues/13314#issue-1119883623>, for anyone wandering the mailing list. - Lucas On Monday, January 31, 2022 at 10:16:42 AM UTC-6 Wolfgang Bangerth wrote: > On 1/28/22 7:50 PM, Lucas Myers wrote: > > ** > >

[deal.II] Re: get_function_values giving different results depending on the number of MPI ranks

e, it might save someone some frustration... In any case, so sorry for all the spam. Kind regards, Lucas On Wednesday, January 26, 2022 at 4:31:21 PM UTC-6 Lucas Myers wrote: > Hi everyone, > > I've been trying to write a parallel distributed solution to a nonlinear > PDE us

[deal.II] get_function_values giving different results depending on the number of MPI ranks

Hi everyone, I've been trying to write a parallel distributed solution to a nonlinear PDE using Newton's method. It runs fine on one rank, but when I use two the solver fails to converge on the second iteration. I've tracked the difference (between one rank and two) down to the `get_function_va

Re: [deal.II] Re: Error loading petsc shared library in step-40 when distributing cell matrix and rhs separately

Matthias, Under /usr the find command doesn't get anything. In fact, if I search for "*petsc*" it only finds the dealii header files. In $HOME it finds: /home/lucas/Application-Data/petsc/src/binding/petsc4py/src/libpetsc4py.h /home/lucas/Application-Data/petsc/src/binding/petsc4py/src/libpetsc

[deal.II] Re: Error loading petsc shared library in step-40 when distributing cell matrix and rhs separately

Bruno, Yes, I'm using bash -- just the default terminal on Ubuntu 20.04. I've attached the output from `make VERBOSE=1` in the `gcc_linker_debug.txt`. As an added twist, when I try to compile with clang 13 (just step-40, not deal.II -- that's still compiled with gcc) it tries to look for a b

[deal.II] Re: Error loading petsc shared library in step-40 when distributing cell matrix and rhs separately

Bruno, You're saying I just need to export the path before running the executable? If that is the case, it is still not working for whatever reason (still giving the linker error). As for two PETSc installations: I vaguely recall trying to install it to root and then quickly trying to undo it

[deal.II] Re: Error loading petsc shared library in step-40 when distributing cell matrix and rhs separately

Bruno, Thanks for the quick response! Yes, I do get the error at runtime and I am running from the terminal. The call in both cases is `mpirun -np 6 ./install/bin/examples/step-40`. The output from ldd is in the attached file. - Lucas On Thursday, January 20, 2022 at 7:54:21 AM UTC-6 bruno.t.

[deal.II] Error loading petsc shared library in step-40 when distributing cell matrix and rhs separately

Hi folks, For debugging purposes, I need to use the `distribute_local_to_global` function which can distribute vectors and matrices separately (this guy in the docs ). However, when I try

[deal.II] Re: Boost version problem

ts to reinstall with a different configuration... - Lucas On Monday, October 25, 2021 at 10:33:07 AM UTC-5 Lucas Myers wrote: > Hi Jean-Paul, > > Sorry to bring this back up again, but I'm running into it after > installing dealii on a new machine -- last time the solution was jus

[deal.II] dealii not compiling because of boost geometry problem

Hi everyone, I'm trying to install dealii on Ubuntu 20.04. For this, I manually installed the latest Boost library first (Boost 1.77.0) as described here . It looks like dealii has found this external boost library,

[deal.II] Re: Boost version problem

know how to tell the compiler where the proper boost installation is. Any help is appreciated. Kind regards, Lucas On Tuesday, August 10, 2021 at 4:49:28 PM UTC-5 Lucas Myers wrote: > Update: the problem is that, when I installed the newest version of > dealii, the boost version

[deal.II] Re: Boost version problem

boost, and then re-install dealii so that it uses its bundled version. Just wanted to make a note about it because I think this is unintended behavior. - Lucas On Tuesday, August 10, 2021 at 3:46:17 PM UTC-5 Lucas Myers wrote: > Hi folks, > > I'm getting the error "The version

[deal.II] Boost version problem

Hi folks, I'm getting the error "The version number of boost that you are compiling with does not match the version number of boost found during deal.II's configuration step." when I try to compile the first tutorial program. It is thrown at /usr/local/include/deal.II/base/config.h:508:17 This

[deal.II] Copying field evaluated at quadrature points to device via CUDA

Hi folks, As I understand it, the standard way of looping through a finite element problem is to iterate through (1) each cell in the mesh, (2) each quadrature point in the cell, (3) the ith shape functions, and (4) the jth shape functions. Given that, would it be possible to copy a finite elem

[deal.II] Re: Able to use umfpack in debug configuration, but not in release configuration

he Release and Debug configurations. Given that the problem was a silly mistake on my end, ought I delete the forum post? Thanks, Lucas On Wednesday, July 21, 2021 at 3:57:12 PM UTC-5 Lucas Myers wrote: > Hi all, > > I've used the #include command in a program > in order to use UMFPack. I

[deal.II] Able to use umfpack in debug configuration, but not in release configuration

Hi all, I've used the #include command in a program in order to use UMFPack. I'm able to compile and run everything fine in Debug mode (provided by Eclipse via the cmake4eclipse add-on). However, when I try to switch to Release mode to optimize the program, it gives me the following warning:

Re: [deal.II] Standard way of applying a function to a finite element vector

software (e.g. Paraview, Visit), or is that done in the C++ source? Just trying to get a sense for best practices... Again, thanks so much for the help! - Lucas On Monday, July 12, 2021 at 11:03:02 PM UTC-5 Wolfgang Bangerth wrote: > On 7/12/21 5:56 PM, Lucas Myers wrote: > > >

[deal.II] Standard way of applying a function to a finite element vector

Hi all, I'd like to apply a method to a finite element function (that is, something that looks like the solution vectors) in order to get another finite element function. For example, suppose my finite element function is a symmetric rank-2 tensor in type, and I would like a finite element func

Re: [deal.II] 5x5 Matrix Inversion Problem

nverses. > > The Solver classes are more than you need here - they are optimized for > large sparse problems, not small dense ones. > > Best, > David Wells > > > -- > *From:* dea...@googlegroups.com on behalf of > Lucas Myers >

[deal.II] 5x5 Matrix Inversion Problem

Hi Everyone, I have a problem wherein one of the terms in my PDE is a 5-component vector function \Lambda of the 5-component solution vector Q. However, I only have an explicit formula for Q as a function of \Lambda and so I am trying to write a numerical inversion class (as a `dealii::Function

[deal.II] Compiling a dealii project without cmake

Hi Everyone, This is sort of a silly question, but how feasible is it to compile and link a dealii project without cmake? I'm pretty new to using C++ so I'm trying to get a handle on the most basic way to do things. I wrote a little file `dealii_function_test.cpp` to test out the Function, Poi