Parallel Dynamic Mesh Generation and Domain Decomposition
نویسندگان
چکیده
The mapping of computations to parallel machines can be realized at the various data structures associated with the computations. In the case of PDE (Partial Differential Equations) based applications, we decided to fonnulate this problem at the discrete data structures of the underlying computation [Chri 91]. In this paper, we fonnulate and implement parallel mapping algorithms on distributed memory machines, including the nCUBE n and Intel iPSC/i860. The uniqueness of OUf parallel mapping scheme is the fact that it integrates mesh splitting (decomposition) with mesh generation. Thus, the mesh generation and meshsplitting preprocessors are integrated into one tool that runs parallel on the targeting machine. Furthermore, the combination of the adaptive mesh refinement and the dynamic domain decomposition becomes· efficient and well defined. Several algorithmic alternatives were investigated for implementing the various parts of this preprocessor [Wu 93-1], [Wu 93-2], including suitable algorithms for mesh generation [Lohn 92], [Khan 91], [Cheng 89] and mesh decomposition [parh 93-1], [Venk 92], [Farh 93-2], [Simon 91J. Local and global mesh refinements are also supported with mesh smoothing and side swapping. Optimal domain partitioning algorithms of the mesh data were considered [Sava 91], [Vand 93]. This work was supported by NSF grants 9123502-CDA and 9202536-CCR. AFOSR F49620-92-J-0069 and PRF grant 6902003.
منابع مشابه
MDEC: MeTiS-based Domain Decomposition for Parallel 2D Mesh Generation
Domain decomposition methods are commonly employed within the context of parallel numerical algorithms. Most often, the domain decomposition is performed before the main computation begins. Within the context of mesh generation, parallel mesh generation is desired when the goal is to mesh a very large geometric domain or if very high accuracy is required. In this paper, we propose a novel techn...
متن کاملA Domain Decomposition Preconditioner for a Parallel Finite Element Solver on Distributed Unstructured Grids
We consider a number of practical issues associated with the parallel distributed memory solution of elliptic partial diierential equations using unstructured meshes in two dimensions. The rst part of the paper describes a parallel mesh generation algorithm which is designed both for eeciency and to produce a well-partitioned, distributed mesh, suitable for the eecient parallel solution of an e...
متن کاملThe hierarchical domain decomposition pre-processing module with the parallel mesh refinement function without communication
This paper describes a parallel fast generation method of large-scale meshes for a hierarchical domain decomposition method implemented in the open source parallel finite element software ADVENTURE. The present authors have newly developed a parallel algorithm such that communication table is updated and inherited in parallel during mesh refinement after the initial communication table is gener...
متن کاملParallel stochastic methods for PDE based grid generation
The efficient generation of meshes is an important step in the numerical solution of various problems in physics and engineering. Here we are interested in situations where global mesh quality and coupling to the physical solution is important. Hence we consider PDE based mesh generation and present a method for the construction of adaptive meshes in two spatial dimensions using domain decompos...
متن کامل