Parallel Methods and Systems for Solving Partial Differential Equations
Abstract
The results of this grant are: (a) development and analysis of new methods and methodologies for solving PDEs on parallel machines, (b) development of mapping strategies of PDE computations to parallel machines, (c) development of knowledge bases for parallel PDE solvers, (d) development of a facility for visualization, collection and analysis of performance data, (e) development of a machine independent object-oriented knowledge interface for specifying PDE computations and solvers, and (f) performance evaluation of PDE solvers on Intel and NCUBE hypercube machines. The feasibility of the proposed ideas was established and usable prototypes have been developed.
Document Details
- Document Type
- Technical Report
- Publication Date
- Apr 30, 1991
- Accession Number
- ADA243822
Entities
People
- Elias N. Houstis
Organizations
- Purdue University