FMDB: Flexible distributed Mesh DataBase for parallel automated adaptive analysis.

摘要

A mesh is piece-wise decomposition of the space/tune domain where used by numerical simulation procedures. The data structure of the mesh strongly influences the overall performance of the simulations since it is an infrastructure executing underneath providing all needed mesh-based operations. From a fact that the flexibility of a mesh data structure comes from the levels of mesh entities and adjacencies present, and by the needs of a distributed mesh data structure operates in a scalable manner, this thesis focused on the development of a Flexible distributed Mesh DataBase (FMDB) capable of shaping its representation based on the specific needs of the application that efficiently supports parallel adaptive analysis in a parallel computing environment.; In order to properly maintain the needed representation even with mesh modification, the mesh entity creation/deletion operators are declared as function objects, initially undetermined. Once the needed representation is provided, they are dynamically set to the proper operators. The needed representation can be provided to the mesh database either by the user explicitly or by running the Dynamic Mesh Usage Monitor, which monitors mesh usage of the application and provides an appropriate representation. For the purpose of supporting distributed meshes on parallel computers, a partition model has been developed. The partition model is located between the partitioned mesh and the geometric model to represent mesh partitioning and support the mesh-based inter-partition operations.; Performance results of the FMDB well demonstrate its efficiency and scalability. Compared to the one-level fixed representation, a decrease in storage obtained with flexible reduced representations is 6--79%. The migration procedure with reduced representations outperforms full representations as the number of entities to migrate or the number of partitions increases. The FMDB is embedded in SCOREC simulation packages effectively supporting parallel automated adaptive analyses.