Something that wasn't obvious to me but is important to point out is that Lizst is typically used with a single, known mesh. The reason Lizst can divide up these meshes is that it compiles for a single mesh at a time, typically preprocessed in advance. It can use the knowledge from properties of the mesh, and the access patterns of different parts of the mesh to figure out how to divide up the meshes in a distributed system, for example.
Something that wasn't obvious to me but is important to point out is that Lizst is typically used with a single, known mesh. The reason Lizst can divide up these meshes is that it compiles for a single mesh at a time, typically preprocessed in advance. It can use the knowledge from properties of the mesh, and the access patterns of different parts of the mesh to figure out how to divide up the meshes in a distributed system, for example.
This comment was marked helpful 0 times.