Abstract
We describe a parallel implementation for large-eddy simulation (LES) of the stratified and rotating turbulence based on MPI. The parallelization strategy is specified by eliminating the tridiagonal solver with explicit method and by a domain decomposition for solving the poisson equation. In this simulation we have run on CRAY-T3E under the message passing platform MPI with various domain decompositions. The processing time and speed up with respect to each domain decomposition and the scalability of this parallel code of LES are presented. The results show that we can gain up to 16 times faster speed up on 64 processors with xyz-direction domain decomposition and scalable up to 128x128x128 which processing time is almost similar to that of 40x40x40 on a single processor machine with a sequential code.
| Original language | English |
|---|---|
| Title of host publication | Proceedings - 4th International Conference/Exhibition on High Performance Computing in the Asia-Pacific Region, HPC-Asia 2000 |
| Publisher | Institute of Electrical and Electronics Engineers Inc. |
| Pages | 884-888 |
| Number of pages | 5 |
| ISBN (Electronic) | 0769505902, 9780769505909 |
| DOIs | |
| Publication status | Published - 2000 |
| Event | 4th International Conference/Exhibition on High Performance Computing in the Asia-Pacific Region, HPC-Asia 2000 - Beijing, China Duration: 2000 May 14 → 2000 May 17 |
Publication series
| Name | Proceedings - 4th International Conference/Exhibition on High Performance Computing in the Asia-Pacific Region, HPC-Asia 2000 |
|---|---|
| Volume | 2 |
Other
| Other | 4th International Conference/Exhibition on High Performance Computing in the Asia-Pacific Region, HPC-Asia 2000 |
|---|---|
| Country/Territory | China |
| City | Beijing |
| Period | 00/5/14 → 00/5/17 |
Bibliographical note
Publisher Copyright:© 2000 IEEE.
All Science Journal Classification (ASJC) codes
- Computational Theory and Mathematics
- Computer Networks and Communications
- Computer Science Applications