<div dir="auto"><div>Interesting, a priori I'm not sure this will work better, mainly because I'd lose the compact band structure.</div><div dir="auto"><br></div><div dir="auto">As for waveform relaxation: I excluded it at first since it appears to be requiring too many CPUs than I have to beat sequential solvers, plus it is more complicated and I have very limited time at this project.</div><div dir="auto"><br></div><div dir="auto">For both suggestions, because of the way the space-time matrix is generated, it is much more convenient for me to mess with the time dimension than with space. </div><div dir="auto"><br></div><div dir="auto">Overall GASM seems a simpler way to go before trying other things.</div><div dir="auto">Please let me know if you decide to add the GASM interfaces.</div><div dir="auto"><br></div><div dir="auto">Thanks again. </div><div dir="auto">Best,</div><div dir="auto">Leonardo</div><div dir="auto"><br><div class="gmail_quote" dir="auto"><div dir="ltr" class="gmail_attr">Il ven 5 mag 2023, 00:01 Matthew Knepley <<a href="mailto:knepley@gmail.com" target="_blank" rel="noreferrer">knepley@gmail.com</a>> ha scritto:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr"><div dir="ltr">On Thu, May 4, 2023 at 1:43 PM LEONARDO MUTTI <<a href="mailto:leonardo.mutti01@universitadipavia.it" rel="noreferrer noreferrer" target="_blank">leonardo.mutti01@universitadipavia.it</a>> wrote:<br></div><div dir="ltr"><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr"><div>Of course, I'll try to explain.</div><div><br></div><div>I am solving a parabolic equation with space-time FEM and I want an efficient solver/preconditioner for the resulting system.</div><div>The corresponding matrix, call it X, has an e.g. block bi-diagonal structure, if the cG(1)-dG(0) method is used (i.e. implicit Euler solved in batch).</div><div>Every block-row of X corresponds to a time instant.</div><div><br></div><div>I want to introduce parallelism in time by subdividing X into overlapping submatrices of e.g 2x2 or 3x3 blocks, along the block diagonal.</div><div>For instance, call X_i the individual blocks. The submatrices would be, for various i, (X_{i-1,i-1},X_{i-1,i};X_{i,i-1},X_{i,i}).</div><div>I'd like each submatrix to be solved in parallel, to combine the various results together in an ASM like fashion.</div><div>Every submatrix has thus a predecessor and a successor, and it overlaps with both, so that as far as I could understand, GASM has to be used in place of ASM.</div></div></blockquote><div><br></div><div>Yes, ordered that way you need GASM. I wonder if inverting the ordering would be useful, namely putting the time index on the inside.</div><div>Then the blocks would be over all time, but limited space, which is more the spirit of ASM I think.</div><div><br></div><div>Have you considered waveform relaxation for this problem?</div><div><br></div><div> Thanks,</div><div><br></div><div> Matt</div><div> <br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr"><div>Hope this helps.</div><div>Best,<br></div><div>Leonardo</div><div><br></div><div class="gmail_quote"><div dir="ltr" class="gmail_attr">Il giorno gio 4 mag 2023 alle ore 18:05 Matthew Knepley <<a href="mailto:knepley@gmail.com" rel="noreferrer noreferrer" target="_blank">knepley@gmail.com</a>> ha scritto:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr"><div dir="ltr">On Thu, May 4, 2023 at 11:24 AM LEONARDO MUTTI <<a href="mailto:leonardo.mutti01@universitadipavia.it" rel="noreferrer noreferrer" target="_blank">leonardo.mutti01@universitadipavia.it</a>> wrote:<br></div><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr">Thank you for the help. <div>Adding to my example:<div><i> call PCGASMSetSubdomains(pc,NSub, subdomains_IS, inflated_IS,ierr)<br> call PCGASMDestroySubdomains(NSub,subdomains_IS,inflated_IS,ierr)<br></i></div><div>results in:<br></div><div><div><i> Error LNK2019 unresolved external symbol PCGASMDESTROYSUBDOMAINS referenced in function ... <br></i></div><div><i> Error LNK2019 unresolved external symbol PCGASMSETSUBDOMAINS referenced in function ... <br></i></div><div>I'm not sure if the interfaces are missing or if I have a compilation problem.</div></div></div></div></blockquote><div><br></div><div>I just want to make sure you really want GASM. It sounded like you might able to do what you want just with ASM.</div><div>Can you tell me again what you want to do overall?</div><div><br></div><div> Thanks,</div><div><br></div><div> Matt</div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr"><div><div>Thank you again. </div><div>Best,<br>Leonardo</div></div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">Il giorno sab 29 apr 2023 alle ore 20:30 Barry Smith <<a href="mailto:bsmith@petsc.dev" rel="noreferrer noreferrer" target="_blank">bsmith@petsc.dev</a>> ha scritto:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div><div><br></div> Thank you for the test code. I have a fix in the branch <span style="color:rgb(51,50,56);font-family:"GitLab Sans",-apple-system,BlinkMacSystemFont,"Segoe UI",Roboto,"Noto Sans",Ubuntu,Cantarell,"Helvetica Neue",sans-serif,"Apple Color Emoji","Segoe UI Emoji","Segoe UI Symbol","Noto Color Emoji";font-size:14px;background-color:rgb(251,250,253)"> </span><a href="https://gitlab.com/petsc/petsc/-/commits/barry/2023-04-29/fix-pcasmcreatesubdomains2d" style="box-sizing:border-box;font-variant-ligatures:none;color:rgb(31,117,203);text-decoration:none;font-size:13.3px" rel="noreferrer noreferrer" target="_blank">barry/2023-04-29/fix-pcasmcreatesubdomains2d</a> with merge request <a href="https://gitlab.com/petsc/petsc/-/merge_requests/6394" rel="noreferrer noreferrer" target="_blank">https://gitlab.com/petsc/petsc/-/merge_requests/6394</a><div><br></div><div> The functions did not have proper Fortran stubs and interfaces so I had to provide them manually in the new branch.<br><div><br></div><div> Use</div><div><br></div><div> git fetch</div><div> git checkout <a href="https://gitlab.com/petsc/petsc/-/commits/barry/2023-04-29/fix-pcasmcreatesubdomains2d" style="box-sizing:border-box;font-variant-ligatures:none;color:rgb(31,117,203);text-decoration:none;font-size:13.3px" rel="noreferrer noreferrer" target="_blank">barry/2023-04-29/fix-pcasmcreatesubdomains2d</a></div><div> ./configure etc</div><div><br></div><div> Your now working test code is in src/ksp/ksp/tests/ex71f.F90 I had to change things slightly and I updated the error handling for the latest version.</div><div><br></div><div> Please let us know if you have any later questions.</div><div><br></div><div> Barry</div><div><br></div><div><br></div><div><br></div><div><div><br><blockquote type="cite"><div>On Apr 28, 2023, at 12:07 PM, LEONARDO MUTTI <<a href="mailto:leonardo.mutti01@universitadipavia.it" rel="noreferrer noreferrer" target="_blank">leonardo.mutti01@universitadipavia.it</a>> wrote:</div><br><div><div dir="ltr">Hello. I am having a hard time understanding the index sets to feed PCGASMSetSubdomains, and I am working in Fortran (as a PETSc novice). To get more intuition on how the IS objects behave I tried the following minimal (non) working example, which should tile a 16x16 matrix into 16 square, non-overlapping submatrices:<div><br></div><div>#include <petsc/finclude/petscmat.h><br></div><div>#include <petsc/finclude/petscksp.h><br>#include <petsc/finclude/petscpc.h><br> USE petscmat<br> USE petscksp<br> USE petscpc<br> <br> Mat :: A<br> PetscInt :: M, NSubx, dof, overlap, NSub<br> INTEGER :: I,J<br> PetscErrorCode :: ierr<br> PetscScalar :: v<br> KSP :: ksp<br> PC :: pc<br> IS :: subdomains_IS, inflated_IS<br> <br> call PetscInitialize(PETSC_NULL_CHARACTER , ierr)</div><div> <br>!-----Create a dummy matrix<br> M = 16<br> call MatCreateAIJ(MPI_COMM_WORLD, PETSC_DECIDE, PETSC_DECIDE,<br> & M, M,<br> & PETSC_DEFAULT_INTEGER, PETSC_NULL_INTEGER,<br> & PETSC_DEFAULT_INTEGER, PETSC_NULL_INTEGER,<br> & A, ierr)<br> <br> DO I=1,M<br> DO J=1,M<br> v = I*J<br> CALL MatSetValue (A,I-1,J-1,v,<br> & INSERT_VALUES , ierr)<br> END DO<br> END DO<br> <br> call MatAssemblyBegin(A,MAT_FINAL_ASSEMBLY , ierr)<br> call MatAssemblyEnd(A,MAT_FINAL_ASSEMBLY , ierr)<br> </div><div>!-----Create KSP and PC<br> call KSPCreate(PETSC_COMM_WORLD,ksp, ierr)<br> call KSPSetOperators(ksp,A,A, ierr)<br> call KSPSetType(ksp,"bcgs",ierr)<br> call KSPGetPC(ksp,pc,ierr)<br> call KSPSetUp(ksp, ierr)<br> call PCSetType(pc,PCGASM, ierr)<br> call PCSetUp(pc , ierr)<br> <br>!-----GASM setup<br> NSubx = 4<br> dof = 1<br> overlap = 0<br> <br> call PCGASMCreateSubdomains2D(pc, <br> & M, M,<br> & NSubx, NSubx,<br> & dof, overlap,<br> & NSub, subdomains_IS, inflated_IS, ierr)<br><br> call ISView(subdomains_IS, PETSC_VIEWER_STDOUT_WORLD, ierr)<br> <br> call KSPDestroy(ksp, ierr) <br> call PetscFinalize(ierr)</div><div><br></div><div>Running this on one processor, I get NSub = 4.</div><div>If PCASM and PCASMCreateSubdomains2D are used instead, I get NSub = 16 as expected.</div><div>Moreover, I get in the end "forrtl: severe (157): Program Exception - access violation". So:</div><div>1) why do I get two different results with ASM, and GASM?</div><div>2) why do I get access violation and how can I solve this?</div><div>In fact, in C, subdomains_IS, inflated_IS should pointers to IS objects. As I see on the Fortran interface, the arguments to PCGASMCreateSubdomains2D are IS objects:<br></div><div><br></div><div> subroutine PCGASMCreateSubdomains2D(a,b,c,d,e,f,g,h,i,j,z)<br> import tPC,tIS<br> PC a ! PC<br> PetscInt b ! PetscInt<br> PetscInt c ! PetscInt<br> PetscInt d ! PetscInt<br> PetscInt e ! PetscInt<br> PetscInt f ! PetscInt<br> PetscInt g ! PetscInt<br> PetscInt h ! PetscInt<br> IS i ! IS<br> IS j ! IS<br> PetscErrorCode z<br> end subroutine PCGASMCreateSubdomains2D<br></div><div>Thus:</div><div>3) what should be inside e.g., subdomains_IS? I expect it to contain, for every created subdomain, the list of rows and columns defining the subblock in the matrix, am I right?</div><div><br></div><div>Context: I have a block-tridiagonal system arising from space-time finite elements, and I want to solve it with GMRES+PCGASM preconditioner, where each overlapping submatrix is on the diagonal and of size 3x3 blocks (and spanning multiple processes). This is PETSc 3.17.1 on Windows.</div><div><br></div><div>Thanks in advance,</div><div>Leonardo</div></div>
</div></blockquote></div><br></div></div></div></blockquote></div>
</blockquote></div><br clear="all"><div><br></div><span>-- </span><br><div dir="ltr"><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div>What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead.<br>-- Norbert Wiener</div><div><br></div><div><a href="http://www.cse.buffalo.edu/~knepley/" rel="noreferrer noreferrer" target="_blank">https://www.cse.buffalo.edu/~knepley/</a><br></div></div></div></div></div></div></div></div>
</blockquote></div></div>
</blockquote></div><br clear="all"><div><br></div><span>-- </span><br><div dir="ltr"><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div>What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead.<br>-- Norbert Wiener</div><div><br></div><div><a href="http://www.cse.buffalo.edu/~knepley/" rel="noreferrer noreferrer" target="_blank">https://www.cse.buffalo.edu/~knepley/</a><br></div></div></div></div></div></div></div></div>
</div>
</blockquote></div></div></div>