- firedrake - imperial.ac.uk

Warning about number of OMP threads
by Andrew Hicks 03 Aug '21

03 Aug '21

Dear Firedrake, I recently tried to run Firedrake using "mpiexec -n 80", so 80 threads, to improve performance. But I got the warning "OMP_NUM_THREADS is not set or is set to a value greater than 1, we suggest setting OMP_NUM_THREADS=1 to improve performance" which doesn't really make sense in my mind since I want 80 threads. Perhaps I am misunderstanding what's going on here. So, is this a problem, and if so, how can I set OMP_NUM_THREADS=1? Andrew Hicks

4 4

Firedrake '21 Online 15-17 September
by Ham, David A 03 Aug '21

03 Aug '21

Dear Firedrakers, The Firedrake ’21 workshop will be held online in the afternoons (UK time) of 15-17 September. This is an opportunity for users and developers of Firedrake and its related tools to present their work, and to learn from and interact with other members of the Firedrake community. I’d like to invite everyone in the Firedrake community, new and old, and no matter what you’re using Firedrake for to come and show off your work. For further details and abstract submission, see: https://firedrakeproject.org/firedrake_21.html . Registration will be just £10 and will open shortly. I hope to see you all there, David -- Dr David Ham Department of Mathematics Imperial College London https://www.imperial.ac.uk/people/david.ham

1 0

installing older version of firedrake
by Miroslav Kuchta 28 Jul '21

28 Jul '21

Hi, is there a recommended way of installing older versions of firedrake? For example, I have previously (some months ago) ran the firedrake-install script on a different system and have the downloaded firedrake folder backed up. Now with a newer version of ubuntu can I somehow reuse the old script/source code downloaded? To provide some context for the question, I am trying to solve Biot equations with stabilized BDM_1-DG_0 elements and observe a different behavior between the versions. However, it's not only firedrake that is different but also the underlying system so I'd like to see where the error comes from. Thanks, Miro

4 5

Problem with Dirichlet BC on edges in a 3D mesh
by Venketeswaran, Abhishek (FN) 19 Jul '21

19 Jul '21

Hi, I created a a couple of physical groups within GMSH with appropriate labels/tags. I then tried to use the same labels after importing the .msh file in firedrake. I got the following error: Please find below the relevant section in the .py file that generates the gmsh mesh and saves it in .msh file gmsh.option.setNumber("Mesh.MshFileVersion", 2) # Physical curves defined with appropriate tags starting from 1000 gmsh.model.addPhysicalGroup(1, b_o_c_arcs,1001) gmsh.model.addPhysicalGroup(1, b_i_c_arcs,1002) gmsh.model.addPhysicalGroup(1, list(t_o_c_arcs),1003) gmsh.model.addPhysicalGroup(1, list(t_i_c_arcs),1004) # Physical surfaces defined with appropriate tags starting from 2000 gmsh.model.addPhysicalGroup(2, circ_pl_surf,2001) gmsh.model.addPhysicalGroup(2, extsurfs,2002) # Physical volumes defined with appropriate tags starting from 3000 gmsh.model.addPhysicalGroup(3, voltags,3001) gmsh.model.mesh.generate(3) gmsh.write("Pipe3D.msh") Here is the snippet of the .py file executed within firedrake: mesh=Mesh('Pipe3D.msh') bc1=DirichletBC(V,zero,1003) When I execute the solve command I get this error: Error: {1003} are not a valid markers (not in {2001,2002}) I think I have a workaround that is not yet giving me errors wherein I define a class derived from DirichletBC and redefine the @utils.cached_property nodes function. However, It would be great if I could do it using the existing DirichletBC class. I would be very grateful if you could help me out here. Thanks -- Abhishek Venketeswaran NETL Research Associate - ORISE National Energy Technology Laboratory Department of Energy abhishek.venketeswaran(a)netl.doe.gov Work: 412-386-4833 Mobile: 716-507-7890 www.netl.doe.gov<http://www.netl.doe.gov> [cid:0311bcdf-830c-456c-a671-ff3e827d18f2]

3 4

Firedrake tutorial on ARCHER2, Online, August 2021
by Betteridge, Jack D 15 Jul '21

15 Jul '21

The Firedrake team will be delivering an online training session on 23rd August 2021, 13:00 - 17:00 BST (12:00 - 16:00 UTC). This course will introduce users to Firedrake and will also be an opportunity to get hands on experience using ARCHER2, the UK national supercomputer. Participants will gain experience writing Firedrake code to solve PDEs, configure solvers and preconditioners for HPC and run the resultant code on ARCHER2. Our course is aimed at researchers who have little or no previous experience of using Firedrake and have a desire to run on HPC. For further details and registration see: https://www.archer2.ac.uk/training/courses/210823-firedrake/

1 0

Building FireDrake + RedHat (Centos,Fedora)
by Régio Pires 14 Jul '21

14 Jul '21

Dear Support, Please, can you help me... FireDrake is compatible with RedHat? Are there a form to build them on this Operation System? (sorry my english) Best Regards -- Régio da Silva Pires Laboratório Nacional de Computação Científica - LNCC regiopires(a)lncc.br 24.2233-6285

2 1

Re: [firedrake] [EXTERNAL] Reg spectral (hp) tetrahedral element
by Lawrence Mitchell 09 Jul '21

09 Jul '21

I realised I forget to leave the mailing list in CC, reintroducing it. > On 9 Jul 2021, at 18:21, Venketeswaran, Abhishek (FN) <Abhishek.Venketeswaran(a)netl.doe.gov> wrote: > > Hi Lawrence, > Thanks for getting back to me. I am currently working on simulation of ultrasonic guided waves on the surface of cylindrical pipes. I am currently using hexahedral elements with GLL quadrature within firedrake to perform this simulation. > Next, I am looking to introduce notch and hole defects on the pipe wall and rerun the simulation. I am not sure on how to create such defects on the pipe wall with the current limitation of defining hexahedral elements using only extrusion of the 2D geometry. I thought, I could circumnavigate this problem by meshing the 3D geometry using tetrahedral elements and then leverage the benefits of spectral element method using the collapsed tetrahedral elements. Kindly let me know how I can solve this problem in Firedrake. Ah, I see. I think that while you can do the 3D meshing with tets, the spectral element stuff is a lot of implementation away from working. If you want to give it a go we can try and provide advice. We're hopeful that unstructured hexahedral meshes will be supported by the end of the year, so that might be swifter! Lawrence

2 3

Reg spectral (hp) tetrahedral element
by Venketeswaran, Abhishek (FN) 08 Jul '21

08 Jul '21

Hi, I would like to use the spectral (hp) tetrahedral elements proposed in the paper below: "A NEW TRIANGULAR AND TETRAHEDRAL BASIS FOR HIGH-ORDER (hp) FINITE ELEMENT METHODS, SPENCER J. SHERWIN AND GEORGE EM KARNIADAKIS, INTERNATIONAL JOURNAL FOR NUMERICAL METHODS IN ENGINEERING, VOL. 38, 3775-3802 (1995) I believe that I most probably saw in the UFL documentation that this element formulation was already implemented. However, I am not able to track that documentation down now. Can you kindly help me with how I may implement this element in Firedrake? Thanks -- Abhishek Venketeswaran NETL Research Associate - ORISE National Energy Technology Laboratory Department of Energy abhishek.venketeswaran(a)netl.doe.gov Work: 412-386-4833 Mobile: 716-507-7890 www.netl.doe.gov<http://www.netl.doe.gov> [cid:9f512d72-18a5-491c-b5bd-71b2da00af09]

1 0

reading group
by Cotter, Colin J 02 Jul '21

02 Jul '21

Dear Firedrakers, Just to remind people that our Firedrake reading group continues on Mondays at 4pm UK time. It's quite a general and very informal reading group on all scientific computing/computational mathematics/numerical analysis topics, just a way to keep us talking together when we can't meet at conferences etc. PhD researchers are especially welcome (and indeed Masters or undergraduate students)! For more details please join #readinggroup on Slack. all the best --cjc Professor Colin Cotter (he/him) Department of Mathematics 755, Huxley Building Imperial College London South Kensington Campus United Kingdom of Great Britain and Northern Ireland +44 2075943468

1 0

Re: [firedrake] Understanding solver parameters in NonlinearVariationalSolver
by Bhavesh Shrimali 29 Jun '21

29 Jun '21

I think I mistakenly clicked reply instead of reply all. Attached is my response: -- Thanks Matt, Thanks for the swift response. These are the logs with basic, backtracking, cp and nleqerr. If I don't specify 'snes_linesearch_type' it defaults to 'basic'. At least that's what I could gather from the log. - `cp` seemingly diverges the fastest. Of course I haven't touched the other options. - `nleqerr` goes the furthest I believe, but not that far from `basic` or `bt`. My concern was that since the forms are essentially the same, and I am passing "seemingly" the same parameters in FEniCS and Firedrake (both of which are using PETSc's SNES), what is it that is different which causes the solver to fail in one but not the other. Edit: Just saw Ed's message. Indeed it defaults to `basic` For definiteness this was the dictionary of solver parameters I am passing to NLVS in Firedrake solver_params={ "snes_type": "newtonls", "ksp_type": "preonly", "pc_type": "lu", "snes_max_it": 100, # "snes_stol": 1.e-6, # "snes_rtol": 1.e-6, # "snes_atol": 1.e-6, "snes_view": None, # "ksp_atol": 1.e-10, # "ksp_rtol":1.e-8, "ksp_view": None, "snes_linesearch_type": "bt", "snes_converged_reason": None, "snes_monitor": None, "ksp_converged_reason": None, "ksp_monitor_true_residual": None } On Sun, 27 Jun 2021 at 11:22, Bhavesh Shrimali <bhavesh.shrimali(a)gmail.com> wrote: > Thanks Matt, > > > Thanks for the swift response. > These are the logs with basic, backtracking, cp and nleqerr. If I don't > specify 'snes_linesearch_type' it defaults to 'basic'. At least that's what > I could gather from the log. > > > - `cp` seemingly diverges the fastest. Of course I haven't touched the > other options. > - `nleqerr` goes the furthest I believe, but not that far from `basic` > or `bt`. > > My concern was that since the forms are essentially the same, and I am > passing "seemingly" the same parameters in FEniCS and Firedrake (both of > which are using PETSc's SNES), what is it that is different which causes > the solver to fail in one but not the other. > > Edit: Just saw Ed's message. Indeed it defaults to `basic` > > > For definiteness this was the dictionary of solver parameters I am passing > to NLVS in Firedrake > > solver_params={ > "snes_type": "newtonls", > "ksp_type": "preonly", > "pc_type": "lu", > "snes_max_it": 100, > # "snes_stol": 1.e-6, > # "snes_rtol": 1.e-6, > # "snes_atol": 1.e-6, > "snes_view": None, > # "ksp_atol": 1.e-10, > # "ksp_rtol":1.e-8, > "ksp_view": None, > "snes_linesearch_type": "bt", > "snes_converged_reason": None, > "snes_monitor": None, > "ksp_converged_reason": None, > "ksp_monitor_true_residual": None > } > > > On Sun, 27 Jun 2021 at 06:00, Matthew Knepley <knepley(a)gmail.com> wrote: > >> Give >> >> -snes_view -snes_converged_reason -snes_monitor -ksp_converged_reason >> -ksp_monitor_true_residual >> >> and send all the output. That way I might be able to see what is >> happening when you do not converge. Also, >> line search 'basic' is not a good choice usually. I would either use the >> default, or 'cp', or maybe 'nleqerr' that >> Patrick added. >> >> Thanks, >> >> Matt >> >> On Sat, Jun 26, 2021 at 10:14 PM Bhavesh Shrimali < >> bhavesh.shrimali(a)gmail.com> wrote: >> >>> Hi everyone, >>> >>> I am trying to understand the usage of solver parameters that are passed >>> to the NonlinearVariationalSolver in firedrake. In order to do some tests, >>> I decided to reproduce some of my previous calculations done in FEniCS for >>> nonlinear viscoelasticity. In old dolfin, I had the following for solver >>> parameters to be passed to PETSc's SNES: >>> >>> snes_solver_parameters = {"nonlinear_solver": "snes", >>> "snes_solver": {"linear_solver": "lu", >>> "maximum_iterations": 50, >>> "report": True, >>> "line_search": 'basic', >>> "error_on_nonconvergence": False, >>> "relative_tolerance": 1e-7, >>> "absolute_tolerance": 1e-8}} >>> ... >>> ... >>> solver = NonlinearVariationalSolver(problem) >>> solver.parameters.update(snes_solver_parameters) >>> >>> Now as I understand, in Firedrake the options passed to the solver are >>> consistent with the corresponding names in PETSc (correct me if I am >>> wrong), and therefore I did: >>> >>> solver_params={ >>> "snes_type": "newtonls", >>> "ksp_type": "preonly", >>> "pc_type": "lu", >>> "snes_max_it": 100, >>> "snes_rtol": 1.e-7, >>> "snes_atol": 1.e-8, >>> "snes_view": None, >>> "ksp_atol": 1.e-10, # not needed? >>> "ksp_rtol":1.e-8, >>> "ksp_view": None, >>> "snes_linesearch_type": "basic", >>> "snes_no_convergence_test": 1 # this has no effect >>> } >>> solver = NonlinearVariationalSolver(problem, >>> solver_parameters=solver_params) >>> >>> However, for some reason, I am unable to get the solver to converge when >>> trying to solve the exact same problem as FEniCS. I experimented with >>> different quadrature degrees as well as tweaked the solver parameters for >>> SNES to best match what I am doing in dolfin. I understand that the form >>> compiler is different hence the precise code that is generated may be >>> different (I am not an expert on code generation) even though the python >>> scripts may look similar due to UFL syntax. >>> >>> In fact the weird part is that the solver seems to converge fine till a >>> certain point in my input (loading). You can see this in the plot below >>> where I plot the final stress-stretch relation obtained at a point in my >>> domain from dolfin and firedrake.. >>> >>> [image: firedrake_vs_fenics.png] >>> >>> >>> I have marked the abscissa beyond which the solver fails to converge in >>> Firedrake. The complete codes are here: >>> https://github.com/bhaveshshrimali/comparingFiredrakeFEniCS >>> >>> I tried setting the option `snes_no_convergence_test` to 1/True but >>> that had no effect as the SNES solver still diverges with `DIVERGED_DTOL`. >>> In the end I put the call to solve inside a try/catch block to at least be >>> able to see the solution till that point. And it surprisingly is accurate >>> till that point. >>> >>> Would anyone have any clue of what could be going wrong? I wouldn't >>> expect that you go through the entire code(s) as that could be fairly long, >>> but any pointers would be helpful :) >>> >>> I am using firedrake via Docker on windows. >>> >>> Thanks >>> >>> -- >>> >>> _______________________________________________ >>> firedrake mailing list >>> firedrake(a)imperial.ac.uk >>> https://mailman.ic.ac.uk/mailman/listinfo/firedrake >>> >> >> >> -- >> What most experimenters take for granted before they begin their >> experiments is infinitely more interesting than any results to which their >> experiments lead. >> -- Norbert Wiener >> >> https://www.cse.buffalo.edu/~knepley/ >> <http://www.cse.buffalo.edu/~knepley/> >> > > > -- > > *Bhavesh Shrimali* > *Ph.D. Student, **Civil and Environmental Engineering,* > > *University of Illinois Urbana-Champaign* > > > *Champaign, IL, United States* > > *Email : bshrima2(a)illinois.edu <bshrima2(a)illinois.edu>* > *Web : **https://bhaveshshrimali.github.io/ > <https://bhaveshshrimali.github.io/>* > -- *Bhavesh Shrimali* *Ph.D. Student, **Civil and Environmental Engineering,* *University of Illinois Urbana-Champaign* *Champaign, IL, United States* *Email : bshrima2(a)illinois.edu <bshrima2(a)illinois.edu>* *Web : **https://bhaveshshrimali.github.io/ <https://bhaveshshrimali.github.io/>*

2 6