{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,2,20]],"date-time":"2025-02-20T05:16:53Z","timestamp":1740028613025,"version":"3.37.3"},"reference-count":0,"publisher":"IOS Press","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2010]]},"abstract":"<jats:p>The Message Passing Interface (MPI) standard defines virtual topologies for optimizing the placement of processes onto processors in order to reduce communication time. That means, processes with their main communication paths represent a graph that has to be cost efficiently mapped onto the actual communication network. In this context, focusing on regular topologies, state-of-the-art mapping strategies that can be applied to any combination of process\/network topology are compared in a two stage approach. First, mapping quality is assessed by a theoretical communication cost measure. Second, based on these results, the most promising methods, which were implemented in Open MPI, are practically compared regarding gain in communication time using MPI's topology mechanism. Finally, a close correspondence between theoretical measure and actual communication time is shown. Additionally, benchmark results prove that optimized process-to-processor mappings can improve communication time by up to 40%, compared to the default linear mapping in many MPI implementations. The findings in this paper can serve as reference not only for MPI implementors, but also for researchers investigating static process-to-processor mappings, in general.<\/jats:p>","DOI":"10.3233\/978-1-60750-530-3-397","type":"book-chapter","created":{"date-parts":[[2025,2,19]],"date-time":"2025-02-19T15:30:51Z","timestamp":1739979051000},"source":"Crossref","is-referenced-by-count":0,"title":["Evaluation of Task Mapping Strategies for Regular Network Topologies"],"prefix":"10.3233","author":[{"family":"Rinke Sebastian","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"family":"Mehlan Torsten","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"family":"Rehm Wolfgang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"7437","container-title":["Advances in Parallel Computing","Parallel Computing: From Multicores and GPU's to Petascale"],"original-title":[],"deposited":{"date-parts":[[2025,2,19]],"date-time":"2025-02-19T15:38:14Z","timestamp":1739979494000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.medra.org\/servlet\/aliasResolver?alias=iospressISSNISBN&issn=0927-5452&volume=19&spage=397"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2010]]},"references-count":0,"URL":"https:\/\/doi.org\/10.3233\/978-1-60750-530-3-397","relation":{},"ISSN":["0927-5452"],"issn-type":[{"value":"0927-5452","type":"print"}],"subject":[],"published":{"date-parts":[[2010]]}}}