Parallel computation of internal and external flows using a portable message-passing harness
Conference
·
OSTI ID:54372
- Daresbury Lab., Warrington (United Kingdom)
The computation of unsteady external aerodynamics and internal chemically reacting flows are of significant importance to a wide range of industrial disciplines. Many of the current solution techniques can be mapped onto the available range of parallel hardware. However, differences in computer hardware, e.g. distributed memory or shared memory, limited the range of architectures the codes will run on. Results for both internal and external flows using implicit and explicit techniques developed within a portable message passing harness are presented. The codes are typical of those encountered in CFD and they can be run on distributed memory machines and shared memory machines with no modifications to the grid allocation strategy. Parallel efficiencies on an Intel i860 64 node hypercube, a 32 node Kendal Square KSR-1 and a range of networked workstations are discussed.
- OSTI ID:
- 54372
- Report Number(s):
- DOE/ER/25151--1-Vol.1; CONF-930331--Vol.1
- Country of Publication:
- United States
- Language:
- English
Similar Records
Message-passing performance of various computers
Global arrays: A portable ``shared-memory`` programming model for distributed memory computers
Global arrays: A portable {open_quotes}shared-memory{close_quotes} programming model for distributed memory computers
Technical Report
·
Wed Jan 31 23:00:00 EST 1996
·
OSTI ID:219525
Global arrays: A portable ``shared-memory`` programming model for distributed memory computers
Conference
·
Fri Dec 30 23:00:00 EST 1994
·
OSTI ID:87640
Global arrays: A portable {open_quotes}shared-memory{close_quotes} programming model for distributed memory computers
Conference
·
Mon Oct 31 23:00:00 EST 1994
·
OSTI ID:86863