This past week, some of us from the Rycroft group attended SIAM CSE 19 at Spokane, WA. One of the presentations I gave is a poster on GPU application performance profiling. This is part of the summer CSGF practicum project I did at LBNL with CCSE group. The applications are written using AMReX framework, and…
Read more Performance study on GPU offloading via CUDA, OpenACC, and OpenMP in AMReX
Multigrid is an iterative method that is particularly suited for solving large, sparse linear systems, in the form of . When we have a sparse matrix at hand, direct methods, such as LU decomposition, for solving the equation becomes undesirable, because we might lose the sparsity and end up with a dense matrix. The best…
Equations for fluid dynamics In one of my research projects, I study numerical methods for fluid-structure interaction (FSI) problems. Examples of this type of problems are ubiquitous in nature, industries, and laboratories. For instance, as organisms swim in a liquid, the parts that generate propulsion, such as flagella and fins, are the solid structures that…
Read more Multigrid applied to Incompressible Navier-Stokes Equation
Many problems in computational sciences require solving linear systems, in the form of . Examples includes integrating an system of ordinary differential equations (ODEs) or solving partial differential equations (PDEs), image processing and analysis, etc. When the problems become large, direct methods, such as LU decomposition, become inefficient. This makes using an iterative method, such…
Read more Hybrid parallelized multigrid with OpenMP, GPU acceleration, and MPI
