Parallelism in solving PDEs (partial differential equations)
This paper examines the potential of parallel computation methods for partial differential equations (PDEs). The author first observes that linear algebra does not give the best data structures for exploiting parllelism in solving PDEs, the data structures should be based on the physical geometry. There is a naturally high level of parallelism in the physical world to be exploited, and he shows there is a natural level of granularity or degree of parallelism that depends on the accuracy needed and the complexity of the PDE problem. This document explores the inherent complexity of parallel methods and parallel machines and conclude that dramatically increased software support is needed for the general scientific and engineering commmunity to exploit the power of highly parallel machines.
- Research Organization:
- Purdue Univ., Lafayette, IN (USA). Dept. of Computer Science
- OSTI ID:
- 7077991
- Report Number(s):
- AD-A-195574/9/XAB
- Country of Publication:
- United States
- Language:
- English
Similar Records
Supercomputers for solving PDE (partial differential equations) problems. Final report, 20 September 1986-11 August 1987
Matrix substructuring, domain decomposition, and particle methods: Current trends for solving PDE's (partial differential equations) in parallel
Related Subjects
71 CLASSICAL AND QUANTUM MECHANICS
GENERAL PHYSICS
PARALLEL PROCESSING
PARTIAL DIFFERENTIAL EQUATIONS
GEOMETRY
NUMERICAL SOLUTION
SUPERCOMPUTERS
COMPUTERS
DIFFERENTIAL EQUATIONS
DIGITAL COMPUTERS
EQUATIONS
MATHEMATICS
PROGRAMMING
990210* - Supercomputers- (1987-1989)
657000 - Theoretical & Mathematical Physics