caa a22 4500 445851996 CHVBK 20180317145416.0 cr unu---uuuuu 170323e20110601xx s 000 0 eng 10.1007/s11227-010-0380-8 doi (NATIONALLICENCE)springer-10.1007/s11227-010-0380-8 Scalable heterogeneous parallelism for atmospheric modeling and simulation [Elektronische Daten] [John Linford, Adrian Sandu] Heterogeneous multicore chipsets with many levels of parallelism are becoming increasingly common in high-performance computing systems. Effective use of parallelism in these new chipsets constitutes the challenge facing a new generation of large scale scientific computing applications. This study examines methods for improving the performance of two-dimensional and three-dimensional atmospheric constituent transport simulation on the Cell Broadband Engine Architecture (CBEA). Afunction offloading approach is used in a 2D transport module, and a vector stream processing approach is used in a 3D transport module. Two methods for transferring incontiguous data between main memory and accelerator local storage are compared. By leveraging the heterogeneous parallelism of the CBEA, the 3D transport module achieves performance comparable to two nodes of an IBM BlueGene/P, or eight Intel Xeon cores, on a single PowerXCell 8i chip. Module performance on two CBEA systems, an IBM BlueGene/P, and an eight-core shared-memory Intel Xeon workstation are given. Springer Science+Business Media, LLC, 2010 Chemical Transport Modeling nationallicence Cell Broadband Engine nationallicence Vector stream processing nationallicence Heterogeneous computing nationallicence Finite difference method nationallicence Fixedgrid nationallicence Linford John Department of Computer Science, Virginia Polytechnic Institute and State University, 2202 Kraft Dr., 24061, Blacksburg, VA, USA aut Sandu Adrian Department of Computer Science, Virginia Polytechnic Institute and State University, 2202 Kraft Dr., 24061, Blacksburg, VA, USA aut The Journal of Supercomputing Springer US; http://www.springer-ny.com 56/3(2011-06-01), 300-327 0920-8542 56:3<300 2011 56 11227 https://doi.org/10.1007/s11227-010-0380-8 text/html Onlinezugriff via DOI 1 research-article jats NATIONALLICENCE 856 40 https://doi.org/10.1007/s11227-010-0380-8 text/html Onlinezugriff via DOI NATIONALLICENCE 700 1- Linford John Department of Computer Science, Virginia Polytechnic Institute and State University, 2202 Kraft Dr., 24061, Blacksburg, VA, USA aut NATIONALLICENCE 700 1- Sandu Adrian Department of Computer Science, Virginia Polytechnic Institute and State University, 2202 Kraft Dr., 24061, Blacksburg, VA, USA aut NATIONALLICENCE 773 0- The Journal of Supercomputing Springer US; http://www.springer-ny.com 56/3(2011-06-01), 300-327 0920-8542 56:3<300 2011 56 11227 Metadata rights reserved Springer special CC-BY-NC licence nationallicence BK010053 XK010053 XK010000 NATIONALLICENCE NATIONALLICENCE NL-springer