Welcome!
To use the personalized features of this site, please log in or register.
If you have forgotten your username or password, we can help.
|
 |
Analyzing the Parallel Scalability of an Implicit Unstructured Mesh CFD Code
| |
|
Analyzing the Parallel Scalability of an Implicit Unstructured Mesh CFD Code
W. D. Gropp7, 8, 9  , D. K. Kaushik7, 8, 9 , D. E. Keyes7, 8, 10 and B. F. Smith7, 8, 9
| (7) |
ISCR, Lawrence Livermore National Laboratory, 94551 Livermore, CA |
| (8) |
ICASE, NASA Langley Research Center, 23681 Hampton, VA |
| (9) |
Math. & Comp. Sci. Division, Argonne National Laboratory, 60439 Argonne, IL |
| (10) |
Math. & Stat. Department, Old Dominion University, 23529 Norfolk, VA |
Abstract
In this paper, we identify the scalability bottlenecks of an unstructured grid CFD code (PETSc-FUN3D) by studying the impact
of several algorithmic and architectural parameters and by examiningdif ferent programmingmodels. We discuss the basic performance
characteristics of this PDE code with the help of simple performance models developed in our earlier work, presentingprimarily
experimental results. In addition to achievingg ood per-processor performance (which has been addressed in our cited work
and without which scalability claims are suspect) we strive to improve the implementation and convergence scalability of PETSc-FUN3D
on thousands of processors.
Fulltext Preview (Small, Large)
 References secured to subscribers.
|
|
|
|
|
|