skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: AMR++: Object-Oriented Parallel Adaptive Mesh Refinement

Technical Report ·
DOI:https://doi.org/10.2172/792630· OSTI ID:792630

Adaptive mesh refinement (AMR) computations are complicated by their dynamic nature. The development of solvers for realistic applications is complicated by both the complexity of the AMR and the geometry of realistic problem domains. The additional complexity of distributed memory parallelism within such AMR applications most commonly exceeds the level of complexity that can be reasonable maintained with traditional approaches toward software development. This paper will present the details of our object-oriented work on the simplification of the use of adaptive mesh refinement on applications with complex geometries for both serial and distributed memory parallel computation. We will present an independent set of object-oriented abstractions (C++ libraries) well suited to the development of such seemingly intractable scientific computations. As an example of the use of this object-oriented approach we will present recent results of an application modeling fluid flow in the eye. Within this example, the geometry is too complicated for a single curvilinear coordinate grid and so a set of overlapping curvilinear coordinate grids' are used. Adaptive mesh refinement and the required grid generation work to support the refinement process is coupled together in the solution of essentially elliptic equations within this domain. This paper will focus on the management of complexity within development of the AMR++ library which forms a part of the Overture object-oriented framework for the solution of partial differential equations within scientific computing.

Research Organization:
Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
Sponsoring Organization:
USDOE Office of Defense Programs (DP) (US)
DOE Contract Number:
W-7405-Eng-48
OSTI ID:
792630
Report Number(s):
UCRL-ID-137373; TRN: US200304%%471
Resource Relation:
Other Information: PBD: 2 Feb 2000
Country of Publication:
United States
Language:
English