[julia-users] Re: Comparison of FE codes; Julia versus commercial FE solver
Hi Petr, In case you weren't aware, there's a Julia package for computation of arbitrary-order Gauss quadrature rules: https://github.com/billmclean/GaussQuadrature.jl -Alex On Tuesday, December 16, 2014 11:31:01 PM UTC-6, Petr Krysl wrote: Hello everybody, In case you're interested, here is the implementation so far. Switch to the top folder, and include one of the examples. If you have paraview, you could check the graphics output (enable graphics export in the source). As always, I'm keen on getting feedback, so don't hesitate please. https://github.com/PetrKryslUCSD/jfineale_for_trying_out Petr
[julia-users] Re: Comparison of FE codes; Julia versus commercial FE solver
Excellent! Missed that one... P On Wednesday, December 17, 2014 1:14:37 PM UTC-8, Alex Ames wrote: Hi Petr, In case you weren't aware, there's a Julia package for computation of arbitrary-order Gauss quadrature rules: https://github.com/billmclean/GaussQuadrature.jl -Alex On Tuesday, December 16, 2014 11:31:01 PM UTC-6, Petr Krysl wrote: Hello everybody, In case you're interested, here is the implementation so far. Switch to the top folder, and include one of the examples. If you have paraview, you could check the graphics output (enable graphics export in the source). As always, I'm keen on getting feedback, so don't hesitate please. https://github.com/PetrKryslUCSD/jfineale_for_trying_out Petr
[julia-users] Re: Comparison of FE codes; Julia versus commercial FE solver
On Wednesday, December 17, 2014 4:14:37 PM UTC-5, Alex Ames wrote: Hi Petr, In case you weren't aware, there's a Julia package for computation of arbitrary-order Gauss quadrature rules: https://github.com/billmclean/GaussQuadrature.jl And there's a somewhat fancier package at: https://github.com/ajt60gaibb/FastGaussQuadrature.jl And if all you want are the plain gauss quadrature rules (no fancy weight functions), you can also use Base.QuadGK.gauss
[julia-users] Re: Comparison of FE codes; Julia versus commercial FE solver
Great. Normally I have no use for Gauss rules beyond order 4 (at most cubic finite elements in the library). But it is good to have access to arbitrary order. Thanks! P On Wednesday, December 17, 2014 1:35:34 PM UTC-8, Steven G. Johnson wrote: In particular, it looks like you just need: param_coords, weights = Base.QuadGK.gauss(Float64, order) in your GaussRule function.
[julia-users] Re: Comparison of FE codes; Julia versus commercial FE solver
Petr, Congratulations for achieving this. Is J FinEALE available somewhere? Maybe as as package? Valentin On Tuesday, 16 December 2014 17:52:29 UTC+1, Petr Krysl wrote: I have made some progress in my effort to gain insight into Julia performance in finite element solvers. I have compared my solver, J FinEALE, (compute and assemble global FE matrices and vectors, then solve the sparse system with Julia's sparse matrix facilities) with the commercial FEA software Comsol 4.4. If anyone is interested, I could document this in a more comprehensive fashion, possibly with the paper. But here goes a real quick rundown: Heat conduction, nonzero essential boundary conditions, nonzero internal heat generation rate. Mesh of 2 million linear triangles. 1,000,000 degrees of freedom. The mesh generation and kickoff of the simulation (assignment of boundary conditions and such) is done by the solver in a separate step, so that is not counted towards the total time spent computing in J FinEALE either. J FinEALE: 20.6 seconds Comsol 4.4 w/ PARDISO: 16 seconds Comsol 4.4 w/ MUMPS: 22 seconds Comsol 4.4 w/ SPOOLES: 37 seconds Just for contrast: Matlab FinEALE: 810 seconds. What I found particularly interesting was what it took to reduce the time for this solution from the initial rewrite from Matlab (around 86 seconds) to the present performance: identify features that degrade performance (declare types!), code critical bottlenecks with ad hoc functions. The critical feature of Julia that made this possible was the speed with which it can execute loops written entirely in Julia. There is no need to call outside help (ala mex-files, which by the way would be of no help in this case). Briefly, Julia DELIVERS! Bravo to the development team. Petr
[julia-users] Re: Comparison of FE codes; Julia versus commercial FE solver
If there proves to be an interest in that, I will certainly make this initial implementation available. Petr On Tuesday, December 16, 2014 9:09:59 AM UTC-8, Valentin Churavy wrote: Petr, Congratulations for achieving this. Is J FinEALE available somewhere? Maybe as as package? Valentin On Tuesday, 16 December 2014 17:52:29 UTC+1, Petr Krysl wrote: I have made some progress in my effort to gain insight into Julia performance in finite element solvers. I have compared my solver, J FinEALE, (compute and assemble global FE matrices and vectors, then solve the sparse system with Julia's sparse matrix facilities) with the commercial FEA software Comsol 4.4. If anyone is interested, I could document this in a more comprehensive fashion, possibly with the paper. But here goes a real quick rundown: Heat conduction, nonzero essential boundary conditions, nonzero internal heat generation rate. Mesh of 2 million linear triangles. 1,000,000 degrees of freedom. The mesh generation and kickoff of the simulation (assignment of boundary conditions and such) is done by the solver in a separate step, so that is not counted towards the total time spent computing in J FinEALE either. J FinEALE: 20.6 seconds Comsol 4.4 w/ PARDISO: 16 seconds Comsol 4.4 w/ MUMPS: 22 seconds Comsol 4.4 w/ SPOOLES: 37 seconds Just for contrast: Matlab FinEALE: 810 seconds. What I found particularly interesting was what it took to reduce the time for this solution from the initial rewrite from Matlab (around 86 seconds) to the present performance: identify features that degrade performance (declare types!), code critical bottlenecks with ad hoc functions. The critical feature of Julia that made this possible was the speed with which it can execute loops written entirely in Julia. There is no need to call outside help (ala mex-files, which by the way would be of no help in this case). Briefly, Julia DELIVERS! Bravo to the development team. Petr
[julia-users] Re: Comparison of FE codes; Julia versus commercial FE solver
I am a COMSOL and Julia user. It will be very interesting to see J FinEALE.
[julia-users] Re: Comparison of FE codes; Julia versus commercial FE solver
Hello everybody, In case you're interested, here is the implementation so far. Switch to the top folder, and include one of the examples. If you have paraview, you could check the graphics output (enable graphics export in the source). As always, I'm keen on getting feedback, so don't hesitate please. https://github.com/PetrKryslUCSD/jfineale_for_trying_out Petr
