<?xml version="1.0" encoding="utf-8"?>
<export-example>
  <doc>
    <id>7230</id>
    <completedYear/>
    <publishedYear>2018</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst>e25851</pageFirst>
    <pageLast/>
    <pageNumber/>
    <edition/>
    <issue>Emerging Architectures in Computational Chemistry</issue>
    <volume/>
    <type>article</type>
    <publisherName>Wiley Online Library</publisherName>
    <publisherPlace/>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>0</belongsToBibliography>
    <completedDate>2018-12-19</completedDate>
    <publishedDate>2018-12-19</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">OpenMP in VASP: Threading and SIMD</title>
    <abstract language="eng">The Vienna Ab initio Simulation Package (VASP) is a widely used electronic structure code that originally exploits process-level parallelism through the Message Passing Interface (MPI) for work distribution within and across nodes.&#13;
Architectural changes of modern parallel processors urge programmers to address thread- and data-level parallelism as well to benefit most from the available compute resources within a node.&#13;
We describe for VASP how to approach for an MPI + OpenMP parallelization including data-level parallelism through OpenMP SIMD constructs together with a generic high-level vector coding scheme.&#13;
We can demonstrate an improved scalability of VASP and more than 20% gain over the MPI-only version, as well as a 2x increased performance of collective operations using the multiple-endpoint MPI feature.&#13;
The high-level vector coding scheme applied to VASP's general gradient approximation routine gives up 9x performance gain on AVX512 platforms with the Intel compiler.</abstract>
    <parentTitle language="deu">International Journal of Quantum Chemistry</parentTitle>
    <identifier type="doi">10.1002/qua.25851</identifier>
    <enrichment key="PeerReviewed">yes</enrichment>
    <enrichment key="SubmissionStatus">in press</enrichment>
    <enrichment key="AcceptedDate">2018-12-19</enrichment>
    <submitter>Florian Wende</submitter>
    <author>Florian Wende</author>
    <author>Martijn Marsman</author>
    <author>Jeongnim Kim</author>
    <author>Fedor Vasilev</author>
    <author>Zhengji Zhao</author>
    <author>Thomas Steinke</author>
    <collection role="institutes" number="vas">Distributed Algorithms and Supercomputing</collection>
    <collection role="persons" number="steinke">Steinke, Thomas</collection>
    <collection role="projects" number="2013-Many-Core-HPC">2013-Many-Core-HPC</collection>
  </doc>
  <doc>
    <id>5947</id>
    <completedYear/>
    <publishedYear>2016</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst/>
    <pageLast/>
    <pageNumber/>
    <edition/>
    <issue/>
    <volume/>
    <type>conferenceobject</type>
    <publisherName/>
    <publisherPlace/>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>0</belongsToBibliography>
    <completedDate>--</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">On Enhancing 3D-FFT Performance in VASP</title>
    <parentTitle language="eng">CUG Proceedings</parentTitle>
    <enrichment key="PeerReviewed">yes</enrichment>
    <enrichment key="FulltextUrl">https://cug.org/proceedings/cug2016_proceedings/includes/files/pap111.pdf</enrichment>
    <author>Florian Wende</author>
    <submitter>Florian Schintke</submitter>
    <author>Martijn Marsman</author>
    <author>Thomas Steinke</author>
    <collection role="institutes" number="vas">Distributed Algorithms and Supercomputing</collection>
    <collection role="persons" number="steinke">Steinke, Thomas</collection>
    <collection role="projects" number="2013-Many-Core-HPC">2013-Many-Core-HPC</collection>
  </doc>
</export-example>
