AuthorsX. Cai
TitleUnstructured computational meshes and data locality
AfilliationScientific Computing
Project(s)Meeting Exascale Computing with Source-to-Source Compilers, Department of High Performance Computing
StatusPublished
Publication TypeTalks, invited
Year of Publication2019
Location of TalkFifth Workshop on Programming Abstractions for Data Locality (PADAL'19), Inria Bordeaux
Keywordsdata locality, HPC
Abstract

Many scientific and engineering applications rely on unstructured computational meshes to capture the irregular shapes and intricate details involved. With respect to software implementation, unstructured meshes require indirectly-indexed, irregular accesses to data arrays. Attaining data locality in the memory hierarchy is thus challenging. This talk touches two related topics. First, we look at the ordering/clustering of entities in an unstructured mesh with respect to cache efficiency. Second, we re-examine the currently widely-used strategy of mesh partitioning, which is based on partitioning a corresponding graph with edge-cut as the optimisation objective. Mismatches between this mainstream methodology of data decomposition and the increasingly heterogeneous computing platforms will be discussed.

Citation Key26983

Contact person