Overall, the GMG method offers a speed-up of 7.1x–9.2x over the PCG method on the same hardware configuration, while also leading to a smaller average residual.

An in-depth analysis of a GPU-based GMG implementation and a comparison against an optimized preconditioned conjugate gradient method shows the GMG method ...

GPU accelerated geometric multigrid method: Comparison with preconditioned conjugate gradient. September 2015. DOI:10.1109/HPEC.2015.7322480. Conference: 2015 ...

An energy efficiency analysis reveals that the GTX 660M and the more powerful Titan cards require a similar amount of energy for running the GMG algorithm: ...

Jan 20, 2021 · The MG-CG solver designed for the block-structured grid is highly efficient and enables large-scale simulations of two-phase flows on GPU based supercomputers.

Jan 20, 2021 · The MG method is one of the most efficient preconditioners to reduce the computational cost and improve the convergence property in multi-scale ...

GPU accelerated geometric multigrid method: Comparison with preconditioned conjugate gradient. Stroia, Iulian, Itu, Lucian, Nita, Cosmin, Lazar, Laszlo ...

We present an efficient, robust and fully GPU-accelerated aggregation-based algebraic multigrid preconditioning technique for the solution of large sparse ...

Extensions and improvements to a deflated preconditioned conjugate gradient technique are considered. In particular, the use of the technique for embedded grids ...

The main kernels of the MG-CG solver achieve more than 90% of the roofline performance. The MG preconditioner is constructed based on the geometric MG method ...