Algorithm 2
From: A multi-GPU enabled solver in Kronecker product form for multiphysics problems

CUDA kernel implementation for \(\mathcal {\hat{X}} {\mathbf {A}}^T\)
From: A multi-GPU enabled solver in Kronecker product form for multiphysics problems

CUDA kernel implementation for \(\mathcal {\hat{X}} {\mathbf {A}}^T\)