Re: [pyfrmailinglist] Nvidia device not detected

Hi Ismael, Thanks for your interest in PyFR. If you run pyfr-sim -h you will get: usage: pyfr-sim [-h] [--verbose] [--backend BACKEND] [--progress] [--nansweep N] {run,restart} ... Runs a PyFR simulation positional arguments: {run,restart} sub-command

Re: [pyfrmailinglist] Setting inflow with a variable expression?

Hi Anthony, Thanks for your interest in PyFR. I'm afraid this is not a feature at the moment - but it is definitely on our todo list! Cheers Peter On 19 Aug 2014, at 11:43, Anthony Ashley ash...@rpi.edumailto:ash...@rpi.edu wrote: Hello, I was wondering if it is possible to specify an

[pyfrmailinglist] PyFR v0.2.2

Dear All, Just to let you know we have released PyFR v0.2.2. PyFR v0.2.2 = PyFR v0.2.1 + mesh/solution re-partitioning + improved device selection for heterogeneous computing + improved documentation. http://www.pyfr.org/download.php Thanks again for your interest in the project. Peter Dr

Re: [pyfrmailinglist] pyfrmailinglist@googlegroups.com

-- Zhang Wei waynezw0...@gmail.commailto:waynezw0...@gmail.com On 17 Aug, 2014, at 11:09 pm, Vincent, Peter E p.vinc...@imperial.ac.uk wrote: Not at the moment I'm afraid. Cheers Peter On 17 Aug 2014, at 08:01, Zhang Wei waynezw0...@gmail.com wrote: Dear Pyfrer: I am very new to pyfr. I want

Re: [pyfrmailinglist] Extrapolate Variables to Flux Points on Faces

Yes, correct. The physical solution is stored at the solution points and extrapolated to the flux points. Cheers Peter Dr Peter Vincent MSci ARCS DIC PhD Senior Lecturer and EPSRC Early Career Fellow Department of Aeronautics Imperial College London South Kensington London SW7 2AZ UK web:

Re: [pyfrmailinglist] OS X PyFR origin/develop branch

Hi Zach, To avoid conflicts between different version of python and other dependencies I would suggest using virtualenv: http://docs.python-guide.org/en/latest/dev/virtualenvs/ I am planning to add something to the User Guide in the near future suggesting the use of virtualenvs. Cheers

Re: [pyfrmailinglist] OS X PyFR origin/develop branch

/. Thanks! make: *** [latexpdf] Error 1 Best Regards, Zach Davis [Rescale] Rescale, Inc. 944 Market Street, Ste. 300 San Francisco, CA 94102 z...@rescale.commailto:z...@rescale.com P: (855) 737-2253 On Mar 9, 2015, at 11:28 AM, Vincent, Peter E p.vinc

[pyfrmailinglist] PyFR v0.8.0

Dear All, We have released PyFR v0.8.0. New Features: - Shock capturing - Parallel .vtu output - Surface-force extraction Download: http://www.pyfr.org/download.php Thanks again for your interest in the project. Peter Dr Peter Vincent MSci ARCS DIC PhD Senior Lecturer and EPSRC

[pyfrmailinglist] PyFR v1.1.0

Dear All, We have released PyFR v1.1.0. New Features: - Residual plugin - More spatially/temporally varying boundary conditions - Viscous stresses from surface force plugin Download: http://www.pyfr.org/download.php Thanks again for your interest in the project. Peter Dr Peter

Re: [pyfrmailinglist] Using PyFR for incompressible boundary layer solution

Hi Dinesh, Thanks for your interest in PyFR. Currently PyFR only has a compressible solver, but we seem to get good agreement with incompressible data by running with Ma = 0.1-0.2. You would probably want to try a characteristic BC (char-riem-inv) at the domain outlet in the first instance.

[pyfrmailinglist] HPCWire Readers Choice Awards 2015

Hi All, Just to let you know that PyFR has been nominated for several HPCWire Readers Choice Awards: 12. Best HPC Software Product or Technology 18. Best HPC Collaboration Between Academia & Industry 20. Top 5 New Products or Technologies to Watch

Re: [pyfrmailinglist] units for mass, length and time

Hi Paul, Thanks for your interest in PyFR. A current feature of PyFR is that it can operate in any consistent system of units. Yes. Would it be possible to add the ability to use a specific system of units; SI, for example? I believe that this capability is a subset of the above? If you

Re: [pyfrmailinglist] cblas openmp error

Hi Emre, Thanks for your interest in PyFR. configparser.NoOptionError: No option 'cblas' in section: 'backend-openmp’ Seems like you have not set cblas in the ini file (as per the User Guide http://www.pyfr.org/user_guide.php) Cheers Peter Dr Peter Vincent MSci ARCS DIC PhD Senior Lecturer

Re: [pyfrmailinglist] minor issue related to [soln-plugin-residual] nstep - pyfr-1.4.0

Hi Piotr, Thanks for the feedback. We can add a note indicating nsteps > 1. But I don’t think nsteps has to be even? Cheers Peter Dr Peter Vincent MSci ARCS DIC PhD Senior Lecturer and EPSRC Early Career Fellow Department of Aeronautics Imperial College London South Kensington London SW7 2AZ

Re: [pyfrmailinglist] pyfr version and the output

Hi Piotr, Thanks for the suggestion. This is something we have discussed quite a lot. One issue is that you can never be sure that the version that is run is exactly the same as a release. The user may have made some ‘small tweaks’ (or even bigger changes), and the code would still tag results

[pyfrmailinglist] PyFR v1.4.0

Dear All, We have released PyFR v1.4.0. New Features: - CGNS mesh importer - MIC backend for Intel Xeon Phi co-processors Download: http://www.pyfr.org/download.php Thanks again for your interest in the project. Peter Dr Peter Vincent MSci ARCS DIC PhD Senior Lecturer and EPSRC Early

Re: [pyfrmailinglist] Minimum step sized rejected and NaNs problem.

Hi Cristale, You only appear to have two P2 elements over the entire jet as it enters the domain - so it is likely under-resolved. You probably need to make the mesh finer there as a starting point. Also, the jet appears to enter parallel and adjacent to a no-slip wall. Is that by design?

Re: [pyfrmailinglist] Re: Multi-GPU per node node runs with PyFR

Hi Robert, Thanks for your interest in PyFR! My last question was a little cheeky I admit. I was curious if the cases from Vermeire, Witherden and Vincent (JCP 2017) paper are available. I can see the zip file among supplementary material but was unable to download it from Elsevier page even

[pyfrmailinglist] PyFR v1.5.0

Dear All, We have released PyFR v1.5.0. New Features: - Dual time-stepping - Support for CUDA aware MPI Download: http://www.pyfr.org/download.php Thanks again for your interest in the project. Peter Dr Peter Vincent MSci ARCS DIC PhD Reader in Aeronautics and EPSRC Fellow Department

Re: [pyfrmailinglist] problem about partitioning mesh file

Do you have scotch or metis installed? Dr Peter Vincent MSci ARCS DIC PhD Reader in Aeronautics and EPSRC Fellow Department of Aeronautics Imperial College London South Kensington London SW7 2AZ email: p.vinc...@imperial.ac.uk web:

Re: [pyfrmailinglist] CUBLAS not Initializing

Many thanks for your interest in the project and the info you have posted. - PyFR 1.50 as of Dec 10, 2016 is NOT ON the Python Package Index (PIP). That is PyFR 1.4. Only two big updates with 1.5 but still! This was my fault - should be up to date now! Thanks! Cheers Peter Dr Peter Vincent

Re: [pyfrmailinglist] Tips on Time Playback of Solution Files

Hi Jonny, Regarding: Interesting! When you say PyFR isn't really used for simulations of sufficient scale, I think you mis-read Freddie’s message, which said: In general, the kinds of simulations that PyFR is used for are of sufficient scale that it is simply not practical to work with more

Re: [pyfrmailinglist] configuration files of your simulation of the test case in the paper: Towards Green Aviation with Python at Petascale

Hi Antonio, The .ini file for the ‘physics run’ is attached. Cheers Peter Dr Peter Vincent MSci ARCS DIC PhD Reader in Aeronautics and EPSRC Fellow Department of Aeronautics Imperial College London South Kensington London SW7 2AZ UK web:

Re: [pyfrmailinglist] Simple testCase Blasius Boundary Layer, just NaNs for rhou,rhov,rhow......

Hi Gabriel, What CFL number are you running at? Peter Dr Peter Vincent MSci ARCS DIC PhD Reader in Aeronautics and EPSRC Fellow Department of Aeronautics Imperial College London South Kensington London SW7 2AZ UK web:

[pyfrmailinglist] PyFR v1.6.0

Dear All, We have released PyFR v1.6.0. New Features: - Added incompressible Euler and Navier-Stokes solvers. - Added support for Intel KNL. - Added libxssm as a matrix multiplication provider for C/OpenMP backend. - Added kernel caching for C/OpenMP backend. Download:

Re: [pyfrmailinglist] Re: How I installed PyFR 1.6.0 with CUDA backend (and dependancies) on Windows 10

Hi Nolan, The (2D) test cases that ship with PyFR are all pathological on terms of their ability to assess performance since they are so small. You should try using one of the bigger 3D cases from one of our recent papers (setup files should be available as supplementary material) Thanks for

Re: [pyfrmailinglist] Options Anti-Aliasing in PyFR

ore. Best regards, Antonio On Tuesday, August 30, 2016 at 11:49:09 AM UTC+2, Vincent, Peter E wrote: Hi Antonio, I have a question about the "Anti-Aliasing" technique available in PyFR in order to improve the stability of the method. I see there are several options availabl

Re: [pyfrmailinglist] Initial condition setting

Hi Junichi, Thanks for trying out PyFR. I set the initial velocity distribution using hyperbolic tangent function, and could run the case. Great! Current issues are shock wave generation from bottom of inlet as shown in the attached figure and termination of simulation job with an error

Re: [pyfrmailinglist] About Standard Scheme for incompressible solver

Hi Will, Thanks for your interest in PyFR. Does it mean that for fully incompressible fluid, artificial compressibility should be "ac-zeta = 0.0” ? No. You need some artificial compressibility for the approach to work. 1. How can I continue my simulation from the latest time step? You can

Re: [pyfrmailinglist] library of 3d examples

Hi Kurt, Thanks for your interest in PyFR. I’m currently working on putting such a library together. In the interim, see the supplementary material of e.g.: http://www.sciencedirect.com/science/article/pii/S0021999116307136 for some 3D test cases. Hope this helps as an interim measure.

[pyfrmailinglist] PyFR v1.7.5

Dear All, We have released PyFR v1.7.5. New Features: - Added support for P0 multi-p convergence acceleration. - Added maximum time-step option for PI controller. Download: http://www.pyfr.org/download.php Thanks again for your interest in the project. Peter Dr Peter Vincent MSci ARCS

[pyfrmailinglist] PyFR v1.8.0

Dear All, We have released PyFR v1.8.0. New Features: - Added locally adaptive pseudo-time-stepping. - Added Riemann invariant boundary condition for artificial compressibility systems. Download: http://www.pyfr.org/download.php Thanks again for your interest in the project. Peter Dr

Re: [pyfrmailinglist] testing PyFR - 3d cylinder - openmp extremely slow

Hi Junting, Thanks for your interest in PyFR. I read that Dr.Vincent said a 3d case setup file was attached to a publication. I could not find it yet. You can find some 3D cases here, for example: https://www.sciencedirect.com/science/article/pii/S0021999116307136 Regards Peter Dr Peter

Re: [pyfrmailinglist] Turbulence models

Hi Douglas PyFR is designed for DNS/ILES so no turbulence models are implemented. Thanks Peter Dr Peter Vincent MSci ARCS DIC PhD FRAeS Reader in Aeronautics and EPSRC Fellow Department of Aeronautics Imperial College London South Kensington London SW7 2AZ UK web:

[pyfrmailinglist] PyFR v1.8.5

Dear All, We have released PyFR v1.8.5. New Features: - Added support for CLBlast. - Added support for Gmsh v4. - Added the optimal Runge-Kutta schemes of Vermeire et al. for pseudo time-stepping. Download: http://www.pyfr.org/download.php Thanks

Re: [pyfrmailinglist] Inlet total pressure

Hi Douglas, Could you send some more details about your setup, and the way in which it is failing? Thanks Peter Dr Peter Vincent MSci ARCS DIC PhD FRAeS Reader in Aeronautics and EPSRC Fellow Department of Aeronautics Imperial College London South Kensington London SW7 2AZ UK web:

[pyfrmailinglist] PyFR Github Organisation

Dear All, PyFR has got its own Github Organisation! The main PyFR repo now lives here: https://github.com/PyFR/PyFR All links to the old repo will automatically redirect, but you may wish to update your Git configuration to avoid any confusion:

Re: [pyfrmailinglist] what would be a good explanation of PyFR having a good utilization of GPU acceleration?

Hi Junting I am trying to find a proper reason for the statement of "PyFR has a good utilization of GPU acceleration technology" , or possibly refer to a paper. I understand that memory transfer between CPU and GPU is one of the big slowdowns of GPU computing. Here is an example of a paper

[pyfrmailinglist] PyFR v1.9.0

Dear All, We have released PyFR v1.9.0. New Features: - Improved strong scaling - Added support for Gmsh v4.1 - Fixed performance issue with OpenCL backend Download: http://www.pyfr.org/download.php Thanks again for your interest in the project. Peter Dr Peter Vincent MSci ARCS DIC PhD

Re: [pyfrmailinglist] WMLES in PyFR

Hi Eduardo, Further to Semih’s reply, I just wanted to pick up on: where I can find LES test cases other than the ones already available in GitHub. The best place to look is in the supplementary material of our publications e.g.

Re: [pyfrmailinglist] sd7003 case with OpenMP

Hi Solal, It looks like you are using the incompressible solver. Was your colleague also using this solver (just so we can be sure that we are comparing apples with apples). Peter Dr Peter Vincent MSci ARCS DIC PhD FRAeS Reader in Aeronautics and EPSRC Fellow Department of Aeronautics

Re: [pyfrmailinglist] sd7003 case with OpenMP

Hi Solal, So as a starting point it might be worth doing a comparison of expected theoretical peak FLOPs and bandwidth between the two platforms (2 x V100s vs 9 x Xeons). There is some information here that might serve as a starting point:

[pyfrmailinglist] PyFR Tutorial Day 2020 and PyFR Symposium 2020

Dear All, I am emailing to advertise two PyFR events that will be held at Imperial College London this summer: - 18th June: PyFR Tutorial Day 2020 (hands on tutorials) - 19th June: PyFR Symposium 2020 (talks from industry partners and the PyFR team) You can register for the PyFR Tutorial Day

Re: [pyfrmailinglist] Order of accuracy analysis in ACM

I did not mess with is the LDG beta and tau I double-checked the source terms several times but it seems okay So I am not sure what could be the problem. Appreciate your help. M Kamra On Tue, Oct 6, 2020 at 23:51 Vincent, Peter E mailto:p.vinc...@imperial.ac.uk>> wrote: Hi M Karma

Re: [pyfrmailinglist] Order of accuracy analysis in ACM

so far, then the source term vanishes and that's what's used in the ini file attached. So I don't think that the source term is the cause of the problem. I removed it now just in case but I got the exact same results. Regards, Mohamed Kamra On Wed, Oct 7, 2020 at 10:48 AM Vincent, P

Re: [pyfrmailinglist] Order of accuracy analysis in ACM

bution of U Numerical solution for Mesh 16x16 , order 4.0 Distribution of U Exact solution for Mesh 16x16 , order 4.0 On Wed, Oct 7, 2020 at 11:28 AM Vincent, Peter E mailto:p.vinc...@imperial.ac.uk>> wrote: So what orders of accuracy do you get? Peter Professor of Computational Fl

Re: [pyfrmailinglist] Order of accuracy analysis in ACM

Hi M Karma, Could you provide more details of the test case, the way your are obtaining the order of accuracy, and the results that you are getting. Thanks Peter Professor of Computational Fluid Dynamics and EPSRC Fellow Department of Aeronautics Imperial College London South Kensington

Re: [pyfrmailinglist] Documentation for PyFR

GitHub/wigging/PyFR/venv/lib/python3.7/site-packages/sphinx/util/compat.py)) make: *** [html] Error 2 Any suggestions on how to build the docs locally? On Thursday, September 24, 2020 at 11:38:19 AM UTC-4 Vincent, Peter E wrote: Hi Gavin, Yes, the html version of the docs are hosted on

Re: [pyfrmailinglist] Documentation for PyFR

wigging/PyFR/venv/lib/python3.7/site-packages/sphinx/util/compat.py)) make: *** [html] Error 2 Any suggestions on how to build the docs locally? On Thursday, September 24, 2020 at 11:38:19 AM UTC-4 Vincent, Peter E wrote: Hi Gavin, Yes, the html version of the docs are hosted on the PyFR web

Re: [pyfrmailinglist] Documentation for PyFR

Hi Gavin, Yes, the html version of the docs are hosted on the PyFR website. Peter Professor of Computational Fluid Dynamics and EPSRC Fellow Department of Aeronautics Imperial College London South Kensington London SW7 2AZ UK On 24 Sep 2020, at 15:45, 'Gavin Wiggins' via PyFR Mailing

[pyfrmailinglist] PyFR v1.10.0

Dear All, We have released PyFR v1.10.0. New Features: - Added support for writing specific regions of the domain to disk - Fixed statistics collection for dual time stepping integrators Download: http://www.pyfr.org/download.php Thanks again for your interest in the project. Peter Dr

Re: [pyfrmailinglist] Question about vector correction function in PyFR.

Hi Kiny, As Freddie says, PyFR only deals with the divergence of these correction functions, and not the vector correction functions themselves. The DG variants are the default coded into PyFR. To try others you could adopt an approach similar to that which Brian adopted here for tensor

Re: [pyfrmailinglist] PyFR Tutorial Day 2020 and PyFR Symposium 2020

slides of the presentations. Thank you, Andy On Wednesday, March 11, 2020 at 8:14:46 AM UTC-4, Vincent, Peter E wrote: Dear All, I am emailing to advertise two PyFR events that will be held at Imperial College London this summer: - 18th June: PyFR Tutorial Day 2020 (hands on tutorials) - 19th June: PyFR Sym