Dependence of the Parallel Performance of the Atmospheric General Circulation Model for the Earth Simulator on Problem Size.
|
Accession number;03A0133874
|
| Title;Dependence of the Parallel Performance of the Atmospheric General Circulation Model for the Earth Simulator on Problem Size. |
| Author;
OHFUCHI W
(Japan Marine Sci. And Technol. Center)
SHINGU S
(Japan Marine Sci. And Technol. Center)
FUCHIGAMI H
(Nec Informatec Systems, Ltd.)
YAMADA M
(Nec Informatec Systems, Ltd.)
|
Journal Title;NEC Res Dev
|
Journal Code:G0138A
|
ISSN:0547-051X
|
|
VOL.44;NO.1;PAGE.99-103(2003)
|
| Figure&Table&Reference;FIG.4, REF.6 |
| Pub. Country;Japan |
| Language;English |
| Abstract;An atmospheric general circulation model (AGCM), named AFES, was extensively optimized for the Earth Simulator (ES), and achieved sustained performance of 26.58TFLOPS or 65 % of the peak performance with the full configuration of the ES under more or less ideal conditions. The sensitivity of AFES's parallel performance on problem size is measured under more practical conditions in this study. Even though the amount of computational operation of Legendre transform increases as O(M3), where M is the truncated wavenumber, while that of physical parameterization does as O(M2), the M dependence of the computational cost does not behave as expected from the amount of computational operation due to the vector efficiency. Some aspects of the cost of communication are also discussed. The results suggest the following. 1) AFES has been developed mainly for super-high resolution, and its coding is very effective only at high resolution. It may be difficult to make AFES very efficient at any, especially low, resolution. 2) In order to maintain high parallel efficiency on the ES, it is essential to employ MPI coding that keeps message size sufficiently large for efficient utilization of the ES's communication ability, and to keep vector length sufficiently large for efficient vector processing. (author abst.) |
|
|
|
Related Articles;
|
|
