https://puma.ub.uni-stuttgart.de/tag/Parallel%20dynamics%20programming%20vectorization%20simulationPUMA RSS feed for /tag/Parallel%20dynamics%20programming%20vectorization%20simulation2026-04-23T08:48:34+02:00https://puma.ub.uni-stuttgart.de/bibtex/2dffb2a37e29a0efe71208a49226516a6/c.niethammerc.niethammer2017-12-12T09:58:14+01:00dynamics highly molecular parallel programming scalable simulation simulations vectorization <span data-person-type="author" class="authorEditorList "><span><span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Wolfgang Eckhardt" itemprop="url" href="/person/1a0dbaf1fbdff3c8d5401f74e8d6a7aad/author/0"><span itemprop="name">W. Eckhardt</span></a></span>, </span><span><span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Alexander Heinecke" itemprop="url" href="/person/1a0dbaf1fbdff3c8d5401f74e8d6a7aad/author/1"><span itemprop="name">A. Heinecke</span></a></span>, </span><span><span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Reinhold Bader" itemprop="url" href="/person/1a0dbaf1fbdff3c8d5401f74e8d6a7aad/author/2"><span itemprop="name">R. Bader</span></a></span>, </span><span><span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Matthias Brehm" itemprop="url" href="/person/1a0dbaf1fbdff3c8d5401f74e8d6a7aad/author/3"><span itemprop="name">M. Brehm</span></a></span>, </span><span><span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Nicolay Hammer" itemprop="url" href="/person/1a0dbaf1fbdff3c8d5401f74e8d6a7aad/author/4"><span itemprop="name">N. Hammer</span></a></span>, </span><span><span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Herbert Huber" itemprop="url" href="/person/1a0dbaf1fbdff3c8d5401f74e8d6a7aad/author/5"><span itemprop="name">H. Huber</span></a></span>, </span><span><span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Hans-Georg Kleinhenz" itemprop="url" href="/person/1a0dbaf1fbdff3c8d5401f74e8d6a7aad/author/6"><span itemprop="name">H. Kleinhenz</span></a></span>, </span><span><span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Jadran Vrabec" itemprop="url" href="/person/1a0dbaf1fbdff3c8d5401f74e8d6a7aad/author/7"><span itemprop="name">J. Vrabec</span></a></span>, </span><span><span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Hans Hasse" itemprop="url" href="/person/1a0dbaf1fbdff3c8d5401f74e8d6a7aad/author/8"><span itemprop="name">H. Hasse</span></a></span>, </span><span><span itemtype="http://schema.org/Person" itemscope="itemscope" itemprop="author"><a title="Martin Horsch" itemprop="url" href="/person/1a0dbaf1fbdff3c8d5401f74e8d6a7aad/author/9"><span itemprop="name">M. Horsch</span></a></span></span> und 5 andere Autor(en). </span><span class="additional-entrytype-information"><em>Seite <span itemprop="pagination">1--12</span>. </em><em><span itemprop="publisher">Springer Berlin Heidelberg</span>, </em><em>Berlin, Heidelberg, </em><span itemtype="http://schema.org/Book" itemscope="itemscope" itemprop="isPartOf"></span>(<em><span>2013<meta content="2013" itemprop="datePublished"/></span></em>)</span>Tue Dec 12 09:58:14 CET 2017Berlin, HeidelbergSupercomputing: 28th International Supercomputing Conference, ISC 2013, Leipzig, Germany, June 16-20, 2013. Proceedings1--12591 TFLOPS Multi-trillion Particles Simulation on SuperMUC2013dynamics highly molecular parallel programming scalable simulation simulations vectorization Anticipating large-scale molecular dynamics simulations (MD) in nano-fluidics, we conduct performance and scalability studies of an optimized version of the code ls1 mardyn. We present our implementation requiring only 32 Bytes per molecule, which allows us to run the, to our knowledge, largest MD simulation to date. Our optimizations tailored to the Intel Sandy Bridge processor are explained, including vectorization as well as shared-memory parallelization to make use of Hyperthreading. Finally we present results for weak and strong scaling experiments on up to 146016 Cores of SuperMUC at the Leibniz Supercomputing Centre, achieving a speed-up of 133k times which corresponds to an absolute performance of 591.2 TFLOPS.