{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,28]],"date-time":"2026-02-28T01:56:19Z","timestamp":1772243779634,"version":"3.50.1"},"reference-count":0,"publisher":"IOS Press","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2016]]},"abstract":"<jats:p>Numerical weather prediction and climate models like COSMO solve a large set of Partial Differential Equations using stencil computations on structured grids. STELLA (Stencil Loop Language) is a DSL in C++ for finite difference methods on structured grids that abstracts the underlying programming model and optimization techniques used for multiple architectures. The STELLA language allows to describe stencils with a unique source code in a concise way, close to the discretized mathematical description of the PDEs, increasing its readability by hiding the complexity of loops and hardware dependent optimizations. STELLA has been used to port the COSMO dynamical core to GPUs while retaining a single source code for multiple computing architectures. We present the STELLA syntax and focus on new, advanced syntax elements for some of the algorithmic motifs present in the dynamical core of COSMO. Namely, the new features improve strong scalability, increase the conciseness of the STELLA DSL, and significantly improve the performance of the COSMO dynamical core. Performance comparisons on GPU are presented.<\/jats:p>","DOI":"10.3233\/978-1-61499-621-7-637","type":"book-chapter","created":{"date-parts":[[2025,2,19]],"date-time":"2025-02-19T10:30:51Z","timestamp":1739961051000},"source":"Crossref","is-referenced-by-count":0,"title":["Porting the COSMO dynamical core to heterogeneous platforms using STELLA Library"],"prefix":"10.3233","author":[{"family":"Osuna Carlos","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"family":"Fuhrer Oliver","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"family":"Gysi Tobias","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"family":"Schulthess Thomas C.","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"7437","container-title":["Advances in Parallel Computing","Parallel Computing: On the Road to Exascale"],"original-title":[],"deposited":{"date-parts":[[2025,2,19]],"date-time":"2025-02-19T10:40:22Z","timestamp":1739961622000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.medra.org\/servlet\/aliasResolver?alias=iospressISBN&isbn=978-1-61499-620-0&spage=637&doi=10.3233\/978-1-61499-621-7-637"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2016]]},"references-count":0,"URL":"https:\/\/doi.org\/10.3233\/978-1-61499-621-7-637","relation":{"is-cited-by":[{"id-type":"doi","id":"10.1007\/s00382-018-4114-6","asserted-by":"object"}]},"ISSN":["0927-5452"],"issn-type":[{"value":"0927-5452","type":"print"}],"subject":[],"published":{"date-parts":[[2016]]}}}