Hello Abbas, Thank you for providing me with the file FindJulia.cmake!
The reason for why CUDA is mandatory in HierBEM is that the numerical quadrature (Sauter quadrature) for double integrals (with singularities) in Galerkin BEM is much more computation intensive than the usually adopted Gauss quadrature for single integrals in FEM. Take discretization of the hypersingular boundary integral operator as an example. For a double integral on a pair of same quadrangles, the order of the Sauter quadrature should be at least 5. Meanwhile, the 4D integration domain is split into 8 parts. Then the total number of quadrature points is 5^4*8=5000. In comparison, for integration on a hexahedra in FEM, Gauss quadrature of order 3 is usually enough and the total number of quadrature points is only 64. Moreover, the integrand function in Galerkin BEM is more complicated than that in FEM. Therefore, HierBEM uses hardware acceleration to make the performance of Galerkin BEM tolerable for engineering cases. Jihuan On Fri, Dec 12, 2025 at 5:23 PM Abbas Ballout <[email protected]> wrote: > Seems cool. > Couldn't build Hier because I don't have Cuda configured yet. > This cmake *might* be useful for you if you are configuring with julia: > https://github.com/JuliaInterop/libcxxwrap-julia/blob/main/FindJulia.cmake > > Abbas > On Monday, December 8, 2025 at 3:05:22 PM UTC+1 Jihuan Tian wrote: > >> Hello everyone! >> >> My name is Jihuan Tian, an independent researcher in Shanghai, China. >> >> I’ve developed HierBEM, a 3D Galerkin boundary element method (BEM) >> library written in C++. Built on deal.II, HierBEM uses hierarchical >> matrices for near log-linear complexity. The numerical quadrature is >> accelerated by CUDA. I hope HierBEM can be a nice supplement to deal.II >> with its capability of handling open domains and directly solving physical >> fields on surfaces with a higher accuracy than FEM. >> >> For more information about HierBEM, please visit the following links: >> >> - Technical features >> <https://jihuan-tian.github.io/hierbem-site/about.html> >> - Several examples with introduction of math theory and explanation >> of source code <https://jihuan-tian.github.io/hierbem-site/examples/> >> - GitHub repository <https://github.com/jihuan-tian/hierbem> >> - Documentation in Doxygen >> <https://jihuan-tian.github.io/hierbem-site/doxygen/index.html> >> >> Feedback and thoughts are very welcome! >> > -- > The deal.II project is located at http://www.dealii.org/ > For mailing list/forum options, see > https://groups.google.com/d/forum/dealii?hl=en > --- > You received this message because you are subscribed to a topic in the > Google Groups "deal.II User Group" group. > To unsubscribe from this topic, visit > https://groups.google.com/d/topic/dealii/T_3LCXoVRA0/unsubscribe. > To unsubscribe from this group and all its topics, send an email to > [email protected]. > To view this discussion visit > https://groups.google.com/d/msgid/dealii/a025f9f5-8748-4a8c-94f9-19cfa4fa1a9en%40googlegroups.com > <https://groups.google.com/d/msgid/dealii/a025f9f5-8748-4a8c-94f9-19cfa4fa1a9en%40googlegroups.com?utm_medium=email&utm_source=footer> > . > -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion visit https://groups.google.com/d/msgid/dealii/CAMNc9yaYkuiVgTLNZgc-GnhmJm9d2TfOF3be9bO8f_ng%2BCo81A%40mail.gmail.com.
