GAMG needs the kernel of the operator to build the coarsening spaces. In elasticity, these are the translation and the rotations. If I were to solve the Stokes problem using the Schur complement approach and I wanted to use GAMG for the Laplacian of the velocity block, should I pass the kernel as well? Is this kernel made of translation and rotation (of the velocity)? I haven’t found anything like this in the literature, hence my question.
Thanks Miguel Miguel A. Salazar de Troya Postdoctoral Researcher, Lawrence Livermore National Laboratory B141 Rm: 1085-5 Ph: 1(925) 422-6411
