S.R.M. Barros, T. Kauranne, Spectral and multigrid spherical Helmholtz equation solvers on distributed memory parallel computers. Proceedings of Fourth Workshop on Use of Parallel Computers in Meteorology, 26--30 Nov, 1990, ECMWF, Reading, UK, 1--27.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....space into 32 processes 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 Fig. 2 Mapping of 32 processes onto processors for 4 Theta 8 partitioning of grid point space. degree of parallelism. The data transposition strategy [1, 4] allows to utilise this approach efficiently on parallel systems. Its idea is to re distribute the complete data to the processes (data transposition) at some stages of the algorithm such that the arithmetic computations between two consecutive transpositions can be performed without any ....

S.R.M. Barros, T. Kauranne, Spectral and multigrid spherical Helmholtz equation solvers on distributed memory parallel computers. Proceedings of Fourth Workshop on Use of Parallel Computers in Meteorology, 26--30 Nov, 1990, ECMWF, Reading, UK, 1--27.

.... fields with spherical basis functions, which have been widely used for modeling phenomena like weather prediction, global warming or global change of atmospheric constituents [2, 3] Spectral models have some advantages over the grid based models commonly parallelized in past and current work[4, 5, 6, 7]. For example, spectral models naturally conserve the area averaged mean square kinetic energy and the mean square vorticity of wind fields, whereas in grid based models these quantities are either not conserved or require additional computation when such conservation is important[8] Despite such ....

.... performed such that the resulting parallel code will perform well even on large scale parallel machines (i.e. machines with a thousand processors) These levels of parallelism are achieved by using an alternative parallelization to the one used in previous work on parallel spectral transport models[6], where parallelism is attained by decomposing atmospheric data along latitudes and or longitudes. In comparison, our approach also takes advantage of the relative independence of computations at different levels in the earth s atmosphere, resulting in parallelism of up to 40 processors, each ....

[Article contains additional citation context not shown here]

S. R. M. Barros and T. Kauranne. Spectral and multigrid spherical helmholtz equation solvers on distributed memory parallel computers. In Fourth Workshop on use of parallel pocessors in meteorology, pages 1--27. European Centre for Mefium-Range Weather Forecasts, November 1990.

....water model) only two transpositions are required, between the grid point space, split in latitudes, and the spectral space, split in wave numbers. For a more complete description of this strategy, and comparisons with the pipeline and 2D static strategies, see works of Barros and Kauranne ([2] and [3] 2.2 The implicit transposition method The use of a virtual shared memory computer frees the programmer from the message passing paradigm. Data needed by several processes can be stored into the shared interleaved memory (SIM) and thus can be addressable by all the processes. On the ....

.... 08 16 20 40 54 80 107 Truncation 106 northern [1 4] 20 [1 8] 10 [1 16] 5 [1 20] 4 [1 40] 2 [1 26] 2 [1 80] 1 [1 80] 1 latitudes [27 54] 1 [81 107] 0 wave [1 3] 27 [1 5] 13 [1 11] 7 [1 13] 5 [1 27] 3 [1] 1 [1 53] 1 [1 107] 1 numbers [4] 26 [6 8] 14 [12 16] 6 [14 20] 6 [28 40] 2 [2 54] 2 [54 80] 2 number of subsets 2 2 2 2 2 3 2 2 Truncation 213 northern [1 8] 20 [1 16] 10 [1 20] 8 [1 40] 4 [1 52] 3 [1 80] 2 [1 53] 2 latitudes [53 54] 2 [54 107] 1 wave [1 6] 27 [1 10] 13 [1 14] 11 [1 26] 5 [1 2] 3 [1 54] 3 [1 107] 2 numbers [7 8] 26 [11 16] 14 [15 20] 10 [27 40] 6 ....

[Article contains additional citation context not shown here]

Saulo R. M. Barros and Tuomo Kauranne. Spectral and Multigrid Spherical Helmholtz Equation Solvers on Distributed Memory Parallel Computers. Proceedings of Fourth Workshop on Use of Parallel Processors in Meteorology, 26-30 November 1990, ECMWF, 1-27.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC