<?xml version="1.0" encoding="utf-8"?>
<export-example>
  <doc>
    <id>5886</id>
    <completedYear/>
    <publishedYear>2025</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst>1777</pageFirst>
    <pageLast>1791</pageLast>
    <pageNumber/>
    <edition/>
    <issue/>
    <volume/>
    <articleNumber/>
    <type>conferenceobject</type>
    <publisherName>ACM</publisherName>
    <publisherPlace>New York</publisherPlace>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>1</belongsToBibliography>
    <completedDate>2025-05-01</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">GPUs All Grown-Up: Fully Device-Driven SpMV Using GPU Work Graphs</title>
    <abstract language="eng">Sparse matrix-vector multiplication (SpMV) is a key operation across high-performance computing, graph analytics, and many more applications. In these applications, the matrix characteristics, notably non-zero elements per row, can vary widely and impact which algorithm performs best. Thus, Graphics Processing Unit (GPU) SpMV algorithms often rely on costly preprocessing to determine what per-row algorithm to select to achieve high performance. In this work we combine SpMV preprocessing and the subsequent per-row processing on the GPU by leveraging the novel “Work Graphs” GPU programming model—initially designed for graphics applications—for dynamic on-device self-scheduling. Work Graphs allow for fine-grain dataflow execution of individual workgroups using emerging hardware and firmware support. As soon as preprocessing has generated sufficient work, workgroups of individual processing kernels are self-scheduled and executed, interleaved with those of other kernels. This improves cache locality and eliminates host interaction altogether. Across a suite of 59 sparse matrices, the best of various novel Work Graphs SpMV implementations outperforms state-of-the-art rocSPARSE “LRB” for a single SpMV by up to 7.19 × (mean: 3.35 ×, SD: 1.89). Furthermore, it achieves much more stable performance across various sparsity patterns than the rocSPARSE CSR-General algorithm, and even beats the advanced rocSPARSE CSR-Adaptive algorithm for up to 92 consecutive SpMV calculations. In addition, compared to rocSPARSE LRB, it reduces code complexity by 75%. Its memory footprint for supporting data structures is a fixed ∼ 25 MiB independent of matrix size, compared to rocSPARSE LRB’s data structures that scale with matrix size to hundreds of megabytes. Overall, this work showcases the performance potential of emerging dynamic on-device scheduling techniques for GPU compute applications.</abstract>
    <parentTitle language="eng">ISCA ´25: Proceedings of the 52nd Annual International Symposium on Computer Architecture</parentTitle>
    <identifier type="isbn">979-8-4007-1261-6</identifier>
    <identifier type="urn">urn:nbn:de:bvb:573-58868</identifier>
    <enrichment key="THI_relatedIdentifier">https://doi.org/10.1145/3695053.3731060</enrichment>
    <enrichment key="THI_review">peer-review</enrichment>
    <enrichment key="THI_openaccess">ja</enrichment>
    <enrichment key="THI_conferenceName">52nd Annual International Symposium on Computer Architecture (ISCA 2025), Tokyo (Japan), 21.-25.06.2025</enrichment>
    <enrichment key="opus.source">publish</enrichment>
    <enrichment key="opus.doi.autoCreate">false</enrichment>
    <enrichment key="opus.urn.autoCreate">true</enrichment>
    <licence>Creative Commons BY-NC-SA 4.0</licence>
    <author>
      <first_name>Fabian</first_name>
      <last_name>Wildgrube</last_name>
    </author>
    <author>
      <first_name>Pete</first_name>
      <last_name>Ehrett</last_name>
    </author>
    <author>
      <first_name>Paul</first_name>
      <last_name>Trojahn</last_name>
    </author>
    <author>
      <first_name>Richard</first_name>
      <last_name>Membarth</last_name>
    </author>
    <author>
      <first_name>Bradford</first_name>
      <last_name>Beckmann</last_name>
    </author>
    <author>
      <first_name>Dominik</first_name>
      <last_name>Baumeister</last_name>
    </author>
    <author>
      <first_name>Matthäus</first_name>
      <last_name>Chajdas</last_name>
    </author>
    <collection role="open_access" number="">open_access</collection>
    <collection role="institutes" number="19311">Fakultät Elektro- und Informationstechnik</collection>
    <collection role="institutes" number="19379">AImotion Bavaria</collection>
    <collection role="persons" number="43674">Membarth, Richard</collection>
    <thesisPublisher>Technische Hochschule Ingolstadt</thesisPublisher>
    <file>https://opus4.kobv.de/opus4-haw/files/5886/3695053.3731060.pdf</file>
  </doc>
  <doc>
    <id>6015</id>
    <completedYear/>
    <publishedYear>2025</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst/>
    <pageLast/>
    <pageNumber>10</pageNumber>
    <edition/>
    <issue/>
    <volume/>
    <articleNumber/>
    <type>conferenceobject</type>
    <publisherName>The Eurographics Association</publisherName>
    <publisherPlace>Goslar</publisherPlace>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>1</belongsToBibliography>
    <completedDate>2025-06-23</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">No More Shading Languages: Compiling C++ to Vulkan Shaders</title>
    <abstract language="eng">Graphics APIs have traditionally relied on shading languages, however, these languages have a number of fundamental defects and limitations. By contrast, GPU compute platforms offer powerful, feature-rich languages suitable for heterogeneous compute. We propose reframing shading languages as embedded domain-specific languages, layered on top of a more general language like C++, doing away with traditional limitations on pointers, functions, and recursion, to the benefit of programmability. This represents a significant compilation challenge because the limitations of shaders are reflected in their lower-level representations. We present the Vcc compiler, which allows conventional C and C++ code to run as Vulkan shaders. Our compiler is complemented by a simple shading library and exposes GPU particulars as intrinsics and annotations. We evaluate the performance of our compiler using a selection of benchmarks, including a real-time path tracer, achieving competitive performance compared to their native CUDA counterparts.</abstract>
    <parentTitle language="eng">High-Performance Graphics - Symposium Papers</parentTitle>
    <identifier type="isbn">978-3-03868-291-2</identifier>
    <identifier type="url">https://diglib.eg.org/handle/10.2312/hpg20251167</identifier>
    <identifier type="urn">urn:nbn:de:bvb:573-60151</identifier>
    <enrichment key="THI_relatedIdentifier">https://doi.org/10.2312/hpg.20251167</enrichment>
    <enrichment key="THI_review">peer-review</enrichment>
    <enrichment key="THI_openaccess">ja</enrichment>
    <enrichment key="THI_conferenceName">High-Performance Graphics (HPG 2025), Copenhagen (Denmark), 23.-25.06.2025</enrichment>
    <enrichment key="opus.source">publish</enrichment>
    <enrichment key="THI_DownloadUrl">https://diglib.eg.org/bitstreams/284640db-e935-40da-afd6-c475e0cb741b/download</enrichment>
    <enrichment key="opus.doi.autoCreate">false</enrichment>
    <enrichment key="opus.urn.autoCreate">true</enrichment>
    <licence>Creative Commons BY 4.0</licence>
    <author>
      <first_name>Hugo</first_name>
      <last_name>Devillers</last_name>
    </author>
    <editor>
      <first_name>Aaron</first_name>
      <last_name>Knoll</last_name>
    </editor>
    <author>
      <first_name>Matthias</first_name>
      <last_name>Kurtenacker</last_name>
    </author>
    <editor>
      <first_name>Christoph</first_name>
      <last_name>Peters</last_name>
    </editor>
    <author>
      <first_name>Richard</first_name>
      <last_name>Membarth</last_name>
    </author>
    <author>
      <first_name>Stefan</first_name>
      <last_name>Lemme</last_name>
    </author>
    <author>
      <first_name>Michael</first_name>
      <last_name>Kenzel</last_name>
    </author>
    <author>
      <first_name>Ömercan</first_name>
      <last_name>Yazici</last_name>
    </author>
    <author>
      <first_name>Philipp</first_name>
      <last_name>Slusallek</last_name>
    </author>
    <collection role="open_access" number="">open_access</collection>
    <collection role="institutes" number="19311">Fakultät Elektro- und Informationstechnik</collection>
    <collection role="institutes" number="19379">AImotion Bavaria</collection>
    <collection role="persons" number="43674">Membarth, Richard</collection>
    <thesisPublisher>Technische Hochschule Ingolstadt</thesisPublisher>
    <file>https://opus4.kobv.de/opus4-haw/files/6015/hpg20251167.pdf</file>
  </doc>
  <doc>
    <id>2877</id>
    <completedYear/>
    <publishedYear>2022</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst/>
    <pageLast/>
    <pageNumber>11</pageNumber>
    <edition/>
    <issue/>
    <volume/>
    <articleNumber>20</articleNumber>
    <type>conferenceobject</type>
    <publisherName>ACM</publisherName>
    <publisherPlace>New York</publisherPlace>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>1</belongsToBibliography>
    <completedDate>2022-10-13</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">AnySeq/GPU: A Novel Approach for Faster Sequence Alignment on GPUs</title>
    <parentTitle language="eng">Proceedings of the 36th ACM International Conference on Supercomputing (ICS-2022)</parentTitle>
    <identifier type="isbn">978-1-4503-9281-5</identifier>
    <enrichment key="THI_review">peer-review</enrichment>
    <enrichment key="THI_openaccess">nein</enrichment>
    <enrichment key="opus.source">publish</enrichment>
    <enrichment key="THI_conferenceName">ICS '22: 2022 International Conference on Supercomputing, online, 28.-30.06.2022</enrichment>
    <enrichment key="THI_relatedIdentifier">https://doi.org/10.1145/3524059.3532376</enrichment>
    <author>
      <first_name>André</first_name>
      <last_name>Müller</last_name>
    </author>
    <author>
      <first_name>Bertil</first_name>
      <last_name>Schmidt</last_name>
    </author>
    <author>
      <first_name>Richard</first_name>
      <last_name>Membarth</last_name>
    </author>
    <author>
      <first_name>Roland</first_name>
      <last_name>Leißa</last_name>
    </author>
    <author>
      <first_name>Sebastian</first_name>
      <last_name>Hack</last_name>
    </author>
    <collection role="institutes" number="19311">Fakultät Elektro- und Informationstechnik</collection>
    <collection role="institutes" number="19379">AImotion Bavaria</collection>
    <collection role="persons" number="43674">Membarth, Richard</collection>
  </doc>
  <doc>
    <id>2025</id>
    <completedYear>2021</completedYear>
    <publishedYear>2021</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst/>
    <pageLast/>
    <pageNumber>11</pageNumber>
    <edition/>
    <issue>54</issue>
    <volume>2021</volume>
    <articleNumber>101425</articleNumber>
    <type>article</type>
    <publisherName>Elsevier</publisherName>
    <publisherPlace>Amsterdam</publisherPlace>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>1</belongsToBibliography>
    <completedDate>2021-07-10</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">tinyMD: Mapping molecular dynamics simulations to heterogeneous hardware using partial evaluation</title>
    <parentTitle language="eng">Journal of Computational Science</parentTitle>
    <identifier type="issn">1877-7503</identifier>
    <enrichment key="opus.source">publish</enrichment>
    <enrichment key="THI_openaccess">nein</enrichment>
    <enrichment key="THI_articleversion">published</enrichment>
    <enrichment key="THI_review">peer-review</enrichment>
    <enrichment key="THI_relatedIdentifier">https://doi.org/10.1016/j.jocs.2021.101425</enrichment>
    <enrichment key="THI_furtherversion">https://opus4.kobv.de/opus4-haw/frontdoor/index/index/docId/4150</enrichment>
    <author>
      <first_name>Rafael</first_name>
      <last_name>Ravedutti Lucio Machado</last_name>
    </author>
    <author>
      <first_name>Jonas</first_name>
      <last_name>Schmitt</last_name>
    </author>
    <author>
      <first_name>Sebastian</first_name>
      <last_name>Eibl</last_name>
    </author>
    <author>
      <first_name>Jan</first_name>
      <last_name>Eitzinger</last_name>
    </author>
    <author>
      <first_name>Roland</first_name>
      <last_name>Leißa</last_name>
    </author>
    <author>
      <first_name>Sebastian</first_name>
      <last_name>Hack</last_name>
    </author>
    <author>
      <first_name>Arsène</first_name>
      <last_name>Pérard-Gayot</last_name>
    </author>
    <author>
      <first_name>Richard</first_name>
      <last_name>Membarth</last_name>
    </author>
    <author>
      <first_name>Harald</first_name>
      <last_name>Köstler</last_name>
    </author>
    <subject>
      <language>deu</language>
      <type>uncontrolled</type>
      <value>Molecular dynamics</value>
    </subject>
    <subject>
      <language>deu</language>
      <type>uncontrolled</type>
      <value>Partial evaluation</value>
    </subject>
    <subject>
      <language>deu</language>
      <type>uncontrolled</type>
      <value>High performance computing</value>
    </subject>
    <subject>
      <language>deu</language>
      <type>uncontrolled</type>
      <value>Load balancing</value>
    </subject>
    <collection role="institutes" number="19311">Fakultät Elektro- und Informationstechnik</collection>
    <collection role="institutes" number="19379">AImotion Bavaria</collection>
    <collection role="persons" number="43674">Membarth, Richard</collection>
  </doc>
  <doc>
    <id>2026</id>
    <completedYear/>
    <publishedYear>2021</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst/>
    <pageLast/>
    <pageNumber>[9]</pageNumber>
    <edition/>
    <issue/>
    <volume/>
    <articleNumber/>
    <type>conferenceobject</type>
    <publisherName>IEEE</publisherName>
    <publisherPlace>Piscataway (NJ)</publisherPlace>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>1</belongsToBibliography>
    <completedDate>2022-03-16</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">FLOWER: a Comprehensive Dataflow Compiler for High-Level Synthesis</title>
    <parentTitle language="eng">Proceedings of the 2021 International Conference on Field-Programmable Technology (ICFPT)</parentTitle>
    <identifier type="isbn">978-1-6654-2010-5</identifier>
    <enrichment key="THI_review">peer-review</enrichment>
    <enrichment key="THI_openaccess">nein</enrichment>
    <enrichment key="opus.source">publish</enrichment>
    <enrichment key="THI_relatedIdentifier">https://doi.org/10.1109/ICFPT52863.2021.9609930</enrichment>
    <enrichment key="THI_conferenceName">2021 International Conference on Field-Programmable Technology (ICFPT), Auckland (Neuseeland), 06.-10. Dezember 2021</enrichment>
    <author>
      <first_name>Puya</first_name>
      <last_name>Amiri</last_name>
    </author>
    <author>
      <first_name>Arsène</first_name>
      <last_name>Pérard-Gayot</last_name>
    </author>
    <author>
      <first_name>Richard</first_name>
      <last_name>Membarth</last_name>
    </author>
    <author>
      <first_name>Philipp</first_name>
      <last_name>Slusallek</last_name>
    </author>
    <author>
      <first_name>Roland</first_name>
      <last_name>Leißa</last_name>
    </author>
    <author>
      <first_name>Sebastian</first_name>
      <last_name>Hack</last_name>
    </author>
    <subject>
      <language>deu</language>
      <type>uncontrolled</type>
      <value>high-level synthesis</value>
    </subject>
    <subject>
      <language>deu</language>
      <type>uncontrolled</type>
      <value>dataflow</value>
    </subject>
    <subject>
      <language>deu</language>
      <type>uncontrolled</type>
      <value>compiler</value>
    </subject>
    <subject>
      <language>deu</language>
      <type>uncontrolled</type>
      <value>FPGA</value>
    </subject>
    <subject>
      <language>deu</language>
      <type>uncontrolled</type>
      <value>transformations</value>
    </subject>
    <subject>
      <language>deu</language>
      <type>uncontrolled</type>
      <value>high-performance computing</value>
    </subject>
    <collection role="institutes" number="19311">Fakultät Elektro- und Informationstechnik</collection>
    <collection role="institutes" number="19379">AImotion Bavaria</collection>
    <collection role="persons" number="43674">Membarth, Richard</collection>
  </doc>
  <doc>
    <id>2028</id>
    <completedYear/>
    <publishedYear>2020</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst>3202</pageFirst>
    <pageLast>3214</pageLast>
    <pageNumber/>
    <edition/>
    <issue>11</issue>
    <volume>39</volume>
    <articleNumber/>
    <type>article</type>
    <publisherName>IEEE</publisherName>
    <publisherPlace>New York</publisherPlace>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>0</belongsToBibliography>
    <completedDate>2022-03-16</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">AnyHLS: High-Level Synthesis with Partial Evaluation</title>
    <parentTitle language="eng">IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (ITCSDI)</parentTitle>
    <identifier type="issn">1937-4151</identifier>
    <enrichment key="THI_relatedIdentifier">https://doi.org/10.1109/TCAD.2020.3012172</enrichment>
    <enrichment key="THI_articleversion">published</enrichment>
    <enrichment key="THI_review">peer-review</enrichment>
    <enrichment key="THI_openaccess">nein</enrichment>
    <enrichment key="opus.source">publish</enrichment>
    <enrichment key="THI_conferenceName">ESWEEK (EMBEDDED SYSTEMS WEEK) 2020, online, 20.-25.09.2020</enrichment>
    <author>
      <first_name>M. Akif</first_name>
      <last_name>Özkan</last_name>
    </author>
    <author>
      <first_name>Arsène</first_name>
      <last_name>Pérard-Gayot</last_name>
    </author>
    <author>
      <first_name>Richard</first_name>
      <last_name>Membarth</last_name>
    </author>
    <author>
      <first_name>Philipp</first_name>
      <last_name>Slusallek</last_name>
    </author>
    <author>
      <first_name>Roland</first_name>
      <last_name>Leißa</last_name>
    </author>
    <author>
      <first_name>Sebastian</first_name>
      <last_name>Hack</last_name>
    </author>
    <author>
      <first_name>Jürgen</first_name>
      <last_name>Teich</last_name>
    </author>
    <author>
      <first_name>Frank</first_name>
      <last_name>Hannig</last_name>
    </author>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Field programmable gate arrays</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>functional programming</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>high level synthesis</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>image processing</value>
    </subject>
    <collection role="persons" number="43674">Membarth, Richard</collection>
  </doc>
  <doc>
    <id>2029</id>
    <completedYear/>
    <publishedYear>2020</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst>1030</pageFirst>
    <pageLast>1040</pageLast>
    <pageNumber/>
    <edition/>
    <issue/>
    <volume/>
    <articleNumber/>
    <type>conferenceobject</type>
    <publisherName>IEEE</publisherName>
    <publisherPlace>Los Alamitos</publisherPlace>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>0</belongsToBibliography>
    <completedDate>2022-03-16</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">AnySeq: A High Performance Sequence Alignment Library based on Partial Evaluation</title>
    <parentTitle language="eng">Proceedings: 2020 IEEE 34th International Parallel and Distributed Processing Symposium (IPDPS 2020)</parentTitle>
    <identifier type="isbn">978-1-7281-6876-0</identifier>
    <identifier type="issn">1530-2075</identifier>
    <enrichment key="THI_relatedIdentifier">https://doi.org/10.1109/IPDPS47924.2020.00109</enrichment>
    <enrichment key="THI_review">peer-review</enrichment>
    <enrichment key="THI_openaccess">nein</enrichment>
    <enrichment key="opus.source">publish</enrichment>
    <enrichment key="THI_conferenceName">2020 IEEE International Parallel and Distributed Processing Symposium (IPDPS), online, 18.-22.05.2020</enrichment>
    <author>
      <first_name>André</first_name>
      <last_name>Müller</last_name>
    </author>
    <author>
      <first_name>Bertil</first_name>
      <last_name>Schmidt</last_name>
    </author>
    <author>
      <first_name>Andreas</first_name>
      <last_name>Hildebrandt</last_name>
    </author>
    <author>
      <first_name>Richard</first_name>
      <last_name>Membarth</last_name>
    </author>
    <author>
      <first_name>Roland</first_name>
      <last_name>Leißa</last_name>
    </author>
    <author>
      <first_name>Matthis</first_name>
      <last_name>Kruse</last_name>
    </author>
    <author>
      <first_name>Sebastian</first_name>
      <last_name>Hack</last_name>
    </author>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Libraries</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Computer architecture</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Generators</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Bioinformatics</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Hardware</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Field programmable gate arrays</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Two dimensional displays</value>
    </subject>
    <collection role="persons" number="43674">Membarth, Richard</collection>
  </doc>
  <doc>
    <id>2030</id>
    <completedYear/>
    <publishedYear>2019</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst/>
    <pageLast/>
    <pageNumber>12</pageNumber>
    <edition/>
    <issue>4</issue>
    <volume>38</volume>
    <articleNumber>40</articleNumber>
    <type>article</type>
    <publisherName>ACM</publisherName>
    <publisherPlace>New York</publisherPlace>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>0</belongsToBibliography>
    <completedDate>2022-03-16</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">Rodent: Generating Renderers without Writing a Generator</title>
    <parentTitle language="eng">ACM Transactions on Graphics (TOG)</parentTitle>
    <identifier type="issn">1557-7368</identifier>
    <identifier type="issn">0730-0301</identifier>
    <enrichment key="THI_relatedIdentifier">https://doi.org/10.1145/3306346.3322955</enrichment>
    <enrichment key="THI_articleversion">published</enrichment>
    <enrichment key="THI_review">peer-review</enrichment>
    <enrichment key="THI_openaccess">nein</enrichment>
    <enrichment key="opus.source">publish</enrichment>
    <enrichment key="THI_conferenceName">SIGGRAPH 2019, Los Angeles (USA), 28.07.-01.08.2019</enrichment>
    <author>
      <first_name>Arsène</first_name>
      <last_name>Pérard-Gayot</last_name>
    </author>
    <author>
      <first_name>Richard</first_name>
      <last_name>Membarth</last_name>
    </author>
    <author>
      <first_name>Roland</first_name>
      <last_name>Leißa</last_name>
    </author>
    <author>
      <first_name>Sebastian</first_name>
      <last_name>Hack</last_name>
    </author>
    <author>
      <first_name>Philipp</first_name>
      <last_name>Slusallek</last_name>
    </author>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Rendering</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Ray-tracing</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Generator</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Partial Evaluation</value>
    </subject>
    <collection role="persons" number="43674">Membarth, Richard</collection>
  </doc>
  <doc>
    <id>2810</id>
    <completedYear/>
    <publishedYear>2018</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst/>
    <pageLast/>
    <pageNumber>30</pageNumber>
    <edition/>
    <issue>OOPSLA</issue>
    <volume>2</volume>
    <articleNumber>119</articleNumber>
    <type>article</type>
    <publisherName>ACM</publisherName>
    <publisherPlace>New York</publisherPlace>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>0</belongsToBibliography>
    <completedDate>2022-09-05</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">AnyDSL: a partial evaluation framework for programming high-performance libraries</title>
    <abstract language="eng">This paper advocates programming high-performance code using partial evaluation. We present a clean-slate programming system with a simple, annotation-based, online partial evaluator that operates on a CPS-style intermediate representation. Our system exposes code generation for accelerators (vectorization/parallelization for CPUs and GPUs) via compiler-known higher-order functions that can be subjected to partial evaluation. This way, generic implementations can be instantiated with target-specific code at compile time.&#13;
&#13;
In our experimental evaluation we present three extensive case studies from image processing, ray tracing, and genome sequence alignment. We demonstrate that using partial evaluation, we obtain high-performance implementations for CPUs and GPUs from one language and one code base in a generic way. The performance of our codes is mostly within 10%, often closer to the performance of multi man-year, industry-grade, manually-optimized expert codes that are considered to be among the top contenders in their fields.</abstract>
    <parentTitle language="eng">Proceedings of the ACM on Programming Languages</parentTitle>
    <identifier type="issn">2475-1421</identifier>
    <identifier type="urn">urn:nbn:de:bvb:573-28104</identifier>
    <enrichment key="THI_relatedIdentifier">https://doi.org/10.1145/3276489</enrichment>
    <enrichment key="THI_review">peer-review</enrichment>
    <enrichment key="THI_openaccess">ja</enrichment>
    <enrichment key="THI_DownloadUrl">https://www.youtube.com/watch?v=pIJga3ioFBs</enrichment>
    <enrichment key="THI_conferenceName">ACM SIGPLAN International Conference on Object-Oriented Programming, Systems, Languages, and Applications: OOPSALA 2018, Boston (USA), 07.11.-09.11.2018</enrichment>
    <enrichment key="opus.source">publish</enrichment>
    <enrichment key="THI_articleversion">published</enrichment>
    <licence>Creative Commons BY-NC-SA 4.0</licence>
    <author>
      <first_name>Roland</first_name>
      <last_name>Leißa</last_name>
    </author>
    <author>
      <first_name>Klaas</first_name>
      <last_name>Boesche</last_name>
    </author>
    <author>
      <first_name>Sebastian</first_name>
      <last_name>Hack</last_name>
    </author>
    <author>
      <first_name>Arsène</first_name>
      <last_name>Pérard-Gayot</last_name>
    </author>
    <author>
      <first_name>Richard</first_name>
      <last_name>Membarth</last_name>
    </author>
    <author>
      <first_name>Philipp</first_name>
      <last_name>Slusallek</last_name>
    </author>
    <author>
      <first_name>André</first_name>
      <last_name>Müller</last_name>
    </author>
    <author>
      <first_name>Bertil</first_name>
      <last_name>Schmidt</last_name>
    </author>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>partial evaluation</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>high-performance</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>parallelization</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>vectorization</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>GPU computing</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>library design</value>
    </subject>
    <collection role="open_access" number="">open_access</collection>
    <collection role="persons" number="43674">Membarth, Richard</collection>
    <thesisPublisher>Technische Hochschule Ingolstadt</thesisPublisher>
    <file>https://opus4.kobv.de/opus4-haw/files/2810/3276489.pdf</file>
  </doc>
  <doc>
    <id>2812</id>
    <completedYear/>
    <publishedYear>2018</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst>17</pageFirst>
    <pageLast>24</pageLast>
    <pageNumber/>
    <edition/>
    <issue/>
    <volume/>
    <articleNumber/>
    <type>conferenceobject</type>
    <publisherName>IEEE</publisherName>
    <publisherPlace>Piscataway (NJ)</publisherPlace>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>0</belongsToBibliography>
    <completedDate>2022-09-05</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">Unified Code Generation for the Parallel Computation of Pairwise Interactions Using Partial Evaluation</title>
    <parentTitle language="eng">Proceedings 17th International Symposium on Parallel and Distributed Computing, ISPDC 2018</parentTitle>
    <identifier type="isbn">978-1-5386-5330-2</identifier>
    <enrichment key="THI_relatedIdentifier">https://doi.org/10.1109/ISPDC2018.2018.00012</enrichment>
    <enrichment key="THI_review">peer-review</enrichment>
    <enrichment key="THI_openaccess">nein</enrichment>
    <enrichment key="THI_conferenceName">17th International Symposium on Parallel and Distributed Computing (ISPDC 2018), Geneva (Switzerland), 25.-28.06.2018</enrichment>
    <enrichment key="opus.source">publish</enrichment>
    <enrichment key="opus.doi.autoCreate">false</enrichment>
    <enrichment key="opus.urn.autoCreate">true</enrichment>
    <author>
      <first_name>Jonas</first_name>
      <last_name>Schmitt</last_name>
    </author>
    <author>
      <first_name>Harald</first_name>
      <last_name>Köstler</last_name>
    </author>
    <author>
      <first_name>Jan</first_name>
      <last_name>Eitzinger</last_name>
    </author>
    <author>
      <first_name>Richard</first_name>
      <last_name>Membarth</last_name>
    </author>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>molecular dynamics</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>code generation</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>partial evaluation</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>gpu</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>neighbor lists</value>
    </subject>
    <collection role="persons" number="43674">Membarth, Richard</collection>
  </doc>
  <doc>
    <id>2998</id>
    <completedYear/>
    <publishedYear>2022</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst/>
    <pageLast/>
    <pageNumber>25</pageNumber>
    <edition/>
    <issue>1</issue>
    <volume>20</volume>
    <articleNumber>17</articleNumber>
    <type>article</type>
    <publisherName>ACM</publisherName>
    <publisherPlace>New York</publisherPlace>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>1</belongsToBibliography>
    <completedDate>2022-12-19</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">XEngine: Optimal Tensor Rematerialization for Neural Networks in Heterogeneous Environments</title>
    <abstract language="eng">Memory efficiency is crucial in training deep learning networks on resource-restricted devices. During backpropagation, forward tensors are used to calculate gradients. Despite the option of keeping those dependencies in memory until they are reused in backpropagation, some forward tensors can be discarded and recomputed later from saved tensors, so-called checkpoints. This allows, in particular, for resource-constrained heterogeneous environments to make use of all available compute devices. Unfortunately, the definition of these checkpoints is a non-trivial problem and poses a challenge to the programmer—improper or excessive recomputations negate the benefit of checkpointing.&#13;
&#13;
In this article, we present XEngine, an approach that schedules network operators to heterogeneous devices in low memory environments by determining checkpoints and recomputations of tensors. Our approach selects suitable resources per timestep and operator and optimizes the end-to-end time for neural networks taking the memory limitation of each device into account. For this, we formulate a mixed-integer quadratic program (MIQP) to schedule operators of deep learning networks on heterogeneous systems. We compare our MIQP solver XEngine against Checkmate [12], a mixed-integer linear programming (MILP) approach that solves recomputation on a single device. Our solver finds solutions that are up to 22.5% faster than the fastest Checkmate schedule in which the network is computed exclusively on a single device. We also find valid schedules for networks making use of both central processing units and graphics processing units if memory limitations do not allow scheduling exclusively to the graphics processing unit.</abstract>
    <parentTitle language="eng">ACM Transactions on Architecture and Code Optimization</parentTitle>
    <identifier type="issn">1544-3973</identifier>
    <identifier type="urn">urn:nbn:de:bvb:573-29989</identifier>
    <enrichment key="THI_relatedIdentifier">https://doi.org/10.1145/3568956</enrichment>
    <enrichment key="THI_articleversion">published</enrichment>
    <enrichment key="THI_review">peer-review</enrichment>
    <enrichment key="THI_openaccess">ja</enrichment>
    <enrichment key="opus.source">publish</enrichment>
    <licence>Creative Commons BY-NC-SA 4.0</licence>
    <author>
      <first_name>Manuela</first_name>
      <last_name>Schuler</last_name>
    </author>
    <author>
      <first_name>Richard</first_name>
      <last_name>Membarth</last_name>
    </author>
    <author>
      <first_name>Philipp</first_name>
      <last_name>Slusallek</last_name>
    </author>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>rematerialization</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>integer linear programming</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>neural networks</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>memory management</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>heterogeneous computing</value>
    </subject>
    <collection role="open_access" number="">open_access</collection>
    <collection role="institutes" number="19311">Fakultät Elektro- und Informationstechnik</collection>
    <collection role="institutes" number="19379">AImotion Bavaria</collection>
    <collection role="persons" number="43674">Membarth, Richard</collection>
    <thesisPublisher>Technische Hochschule Ingolstadt</thesisPublisher>
    <file>https://opus4.kobv.de/opus4-haw/files/2998/3568956.pdf</file>
  </doc>
  <doc>
    <id>3885</id>
    <completedYear/>
    <publishedYear>2023</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst>736</pageFirst>
    <pageLast>745</pageLast>
    <pageNumber/>
    <edition/>
    <issue/>
    <volume/>
    <articleNumber/>
    <type>conferenceobject</type>
    <publisherName>IEEE</publisherName>
    <publisherPlace>Piscataway</publisherPlace>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>1</belongsToBibliography>
    <completedDate>--</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">AnyQ: An Evaluation Framework for Massively-Parallel Queue Algorithms</title>
    <parentTitle language="eng">Proceedings: 2023 IEEE International Parallel and Distributed Processing Symposium: IPDPS 2023</parentTitle>
    <identifier type="isbn">979-8-3503-3766-2</identifier>
    <enrichment key="opus_doi_flag">true</enrichment>
    <enrichment key="opus_import_data">{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2023,8,8]],"date-time":"2023-08-08T04:12:33Z","timestamp":1691467953192},"reference-count":27,"publisher":"IEEE","license":[{"start":{"date-parts":[[2023,5,1]],"date-time":"2023-05-01T00:00:00Z","timestamp":1682899200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2023,5,1]],"date-time":"2023-05-01T00:00:00Z","timestamp":1682899200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/100009950","name":"Ministry of Education","doi-asserted-by":"publisher"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2023,5]]},"DOI":"10.1109\/ipdps54959.2023.00079","type":"proceedings-article","created":{"date-parts":[[2023,7,18]],"date-time":"2023-07-18T17:29:11Z","timestamp":1689701351000},"source":"Crossref","is-referenced-by-count":0,"title":["AnyQ: An Evaluation Framework for Massively-Parallel Queue Algorithms"],"prefix":"10.1109","author":[{"given":"Michael","family":"Kenzel","sequence":"first","affiliation":[{"name":"Deutsches Forschungszentrum f&amp;#x00FC;r K&amp;#x00FC;nstliche Intelligenz (DFKI),Saarland Informatics Campus"}]},{"given":"Stefan","family":"Lemme","sequence":"additional","affiliation":[{"name":"Deutsches Forschungszentrum f&amp;#x00FC;r K&amp;#x00FC;nstliche Intelligenz (DFKI),Saarland Informatics Campus"}]},{"given":"Richard","family":"Membarth","sequence":"additional","affiliation":[{"name":"Deutsches Forschungszentrum f&amp;#x00FC;r K&amp;#x00FC;nstliche Intelligenz (DFKI),Saarland Informatics Campus"}]},{"given":"Matthias","family":"Kurtenacker","sequence":"additional","affiliation":[{"name":"Deutsches Forschungszentrum f&amp;#x00FC;r K&amp;#x00FC;nstliche Intelligenz (DFKI),Saarland Informatics Campus"}]},{"given":"Hugo","family":"Devillers","sequence":"additional","affiliation":[{"name":"Saarland University,Saarland Informatics Campus"}]},{"given":"Markus","family":"Steinberger","sequence":"additional","affiliation":[{"name":"Graz University of Technology"}]},{"given":"Philipp","family":"Slusallek","sequence":"additional","affiliation":[{"name":"Deutsches Forschungszentrum f&amp;#x00FC;r K&amp;#x00FC;nstliche Intelligenz (DFKI),Saarland Informatics Campus"}]}],"member":"263","reference":[{"key":"ref13","article-title":"A new approach to exclusive data access in shared memory multiprocessors","author":"jensen","year":"1987","journal-title":"Technical Report UCRL-53745 Lawrence Livermore National Laboratory"},{"key":"ref12","article-title":"A scalable, portable, and memory-efficient lock-free FIFO queue","author":"nikolaev","year":"2019","journal-title":"CoRR"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/CGO.2004.1281665"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1145\/2668930.2688048"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1145\/319566.319567"},{"key":"ref10","first-page":"267","article-title":"Simple, fast, and practical non-blocking and blocking concurrent queue algorithms","author":"michael","year":"1996","journal-title":"Proc 15th Annual ACM Symposium on Principles of Distributed Computing ser PODC"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1145\/2517327.2442527"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1145\/3205289.3205291"},{"key":"ref17","year":"2022","journal-title":"CUDA C Programming Guide 6 0"},{"key":"ref16","first-page":"119:1","article-title":"AnyDSL: A partial evaluation framework for programming high-performance libraries","volume":"2","author":"lei\u00dfa","year":"2018","journal-title":"Proceedings of the ACM on Programming Languages (PACMPL)"},{"key":"ref19","year":"2018","journal-title":"Hsa programmer reference manual specification 1 1"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/MCSE.2010.69"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.18637\/jss.v008.i14"},{"key":"ref23","year":"2021","journal-title":"Threading Building Blocks"},{"key":"ref26","year":"2022","journal-title":"ARM Architecture Reference Manual ARMv 8 for ARMv8-A Architecture Profile"},{"key":"ref25","article-title":"A fast general purpose lock-free queue for C++","author":"desrochers","year":"2014","journal-title":"moodycamel Tech Rep"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1145\/3192366.3192413"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1145\/1375581.1375591"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/InPar.2012.6339601"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/ISORC.2010.10"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-540-77096-1_29"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-39958-9_18"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/ICDCS.2003.1203503"},{"key":"ref4","article-title":"NVIDIA tesla V100 GPU architecture","year":"2017","journal-title":"NVIDIA Corporation Tech Rep"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1145\/2851141.2851168"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-17653-1_29"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1145\/78969.78972"}],"event":{"name":"2023 IEEE International Parallel and Distributed Processing Symposium (IPDPS)","location":"St. Petersburg, FL, USA","start":{"date-parts":[[2023,5,15]]},"end":{"date-parts":[[2023,5,19]]}},"container-title":["2023 IEEE International Parallel and Distributed Processing Symposium (IPDPS)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/10177277\/10177383\/10177434.pdf?arnumber=10177434","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,8,7]],"date-time":"2023-08-07T17:37:44Z","timestamp":1691429864000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10177434\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,5]]},"references-count":27,"URL":"http:\/\/dx.doi.org\/10.1109\/ipdps54959.2023.00079","relation":{},"published":{"date-parts":[[2023,5]]}}}</enrichment>
    <enrichment key="local_crossrefDocumentType">proceedings-article</enrichment>
    <enrichment key="local_crossrefLicence">https://doi.org/10.15223/policy-029</enrichment>
    <enrichment key="local_import_origin">crossref</enrichment>
    <enrichment key="local_doiImportPopulated">PersonAuthorFirstName_1,PersonAuthorLastName_1,PersonAuthorFirstName_2,PersonAuthorLastName_2,PersonAuthorFirstName_3,PersonAuthorLastName_3,PersonAuthorFirstName_4,PersonAuthorLastName_4,PersonAuthorFirstName_5,PersonAuthorLastName_5,PersonAuthorFirstName_6,PersonAuthorLastName_6,PersonAuthorFirstName_7,PersonAuthorLastName_7,Enrichmentconference_title,Enrichmentconference_place,PublisherName,TitleMain_1,TitleParent_1,CompletedYear,Enrichmentlocal_crossrefLicence</enrichment>
    <enrichment key="conference_title">2023 IEEE International Parallel and Distributed Processing Symposium (IPDPS)</enrichment>
    <enrichment key="conference_place">St. Petersburg, FL, USA</enrichment>
    <enrichment key="opus.source">doi-import</enrichment>
    <enrichment key="THI_relatedIdentifier">https://doi.org/10.1109/IPDPS54959.2023.00079</enrichment>
    <enrichment key="THI_conferenceName">2023 IEEE International Parallel and Distributed Processing Symposium (IPDPS), St. Petersburg (USA), 15.-19.05.2023</enrichment>
    <enrichment key="THI_openaccess">nein</enrichment>
    <enrichment key="THI_review">peer-review</enrichment>
    <enrichment key="opus.doi.autoCreate">false</enrichment>
    <enrichment key="opus.urn.autoCreate">true</enrichment>
    <author>
      <first_name>Michael</first_name>
      <last_name>Kenzel</last_name>
    </author>
    <author>
      <first_name>Stefan</first_name>
      <last_name>Lemme</last_name>
    </author>
    <author>
      <first_name>Richard</first_name>
      <last_name>Membarth</last_name>
    </author>
    <author>
      <first_name>Matthias</first_name>
      <last_name>Kurtenacker</last_name>
    </author>
    <author>
      <first_name>Hugo</first_name>
      <last_name>Devillers</last_name>
    </author>
    <author>
      <first_name>Markus</first_name>
      <last_name>Steinberger</last_name>
    </author>
    <author>
      <first_name>Philipp</first_name>
      <last_name>Slusallek</last_name>
    </author>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Massively Parallel</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Concurrent Queue</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>GPU</value>
    </subject>
    <collection role="institutes" number="19311">Fakultät Elektro- und Informationstechnik</collection>
    <collection role="institutes" number="19379">AImotion Bavaria</collection>
    <collection role="persons" number="43674">Membarth, Richard</collection>
  </doc>
  <doc>
    <id>3115</id>
    <completedYear/>
    <publishedYear>2022</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst>1489</pageFirst>
    <pageLast>1501</pageLast>
    <pageNumber/>
    <edition/>
    <issue>2</issue>
    <volume>30</volume>
    <articleNumber/>
    <type>article</type>
    <publisherName>IEEE</publisherName>
    <publisherPlace>Piscataway</publisherPlace>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>1</belongsToBibliography>
    <completedDate>--</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">Temporal Coherence-Based Distributed Ray Tracing of Massive Scenes</title>
    <parentTitle language="eng">IEEE Transactions on Visualization and Computer Graphics</parentTitle>
    <identifier type="issn">1941-0506</identifier>
    <identifier type="issn">1077-2626</identifier>
    <enrichment key="THI_articleversion">published</enrichment>
    <enrichment key="THI_openaccess">nein</enrichment>
    <enrichment key="THI_relatedIdentifier">https://doi.org/10.1109/TVCG.2022.3219982</enrichment>
    <enrichment key="THI_review">peer-review</enrichment>
    <author>
      <first_name>Xiang</first_name>
      <last_name>Xu</last_name>
    </author>
    <author>
      <first_name>Lu</first_name>
      <last_name>Wang</last_name>
    </author>
    <author>
      <first_name>Arsène</first_name>
      <last_name>Pérard-Gayot</last_name>
    </author>
    <author>
      <first_name>Richard</first_name>
      <last_name>Membarth</last_name>
    </author>
    <author>
      <first_name>Cuiyu</first_name>
      <last_name>Li</last_name>
    </author>
    <author>
      <first_name>Chenglei</first_name>
      <last_name>Yang</last_name>
    </author>
    <author>
      <first_name>Philipp</first_name>
      <last_name>Slusallek</last_name>
    </author>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Computer Graphics</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Ray Tracing</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Distributed Graphics</value>
    </subject>
    <collection role="institutes" number="19311">Fakultät Elektro- und Informationstechnik</collection>
    <collection role="institutes" number="19379">AImotion Bavaria</collection>
    <collection role="persons" number="43674">Membarth, Richard</collection>
  </doc>
  <doc>
    <id>5465</id>
    <completedYear/>
    <publishedYear>2025</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst>51</pageFirst>
    <pageLast>61</pageLast>
    <pageNumber/>
    <edition/>
    <issue/>
    <volume>37</volume>
    <articleNumber/>
    <type>article</type>
    <publisherName>Gesellschaft für Informatik</publisherName>
    <publisherPlace>Bonn</publisherPlace>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>1</belongsToBibliography>
    <completedDate>2024-12-12</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">Exploration of Efficient Computation for Trajectory Planning via Fixed-Point Arithmetic</title>
    <parentTitle language="deu">PARS-Mitteilungen 2025</parentTitle>
    <identifier type="issn">0177-0454</identifier>
    <identifier type="url">https://dl.gi.de/items/6b2c0546-1081-40ab-b97b-052b60b8374f</identifier>
    <enrichment key="THI_articleversion">published</enrichment>
    <enrichment key="THI_review">peer-review</enrichment>
    <enrichment key="THI_openaccess">nein</enrichment>
    <enrichment key="opus.source">publish</enrichment>
    <enrichment key="THI_conferenceName">30. PARS-Workshop, Ingolstadt (Germany), 25.-26.09.2024</enrichment>
    <author>
      <first_name>Wenguang</first_name>
      <last_name>Xu</last_name>
    </author>
    <author>
      <first_name>Richard</first_name>
      <last_name>Membarth</last_name>
    </author>
    <collection role="institutes" number="19311">Fakultät Elektro- und Informationstechnik</collection>
    <collection role="institutes" number="19379">AImotion Bavaria</collection>
    <collection role="persons" number="43674">Membarth, Richard</collection>
  </doc>
  <doc>
    <id>5466</id>
    <completedYear/>
    <publishedYear>2025</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst>63</pageFirst>
    <pageLast>70</pageLast>
    <pageNumber/>
    <edition/>
    <issue/>
    <volume>37</volume>
    <articleNumber/>
    <type>article</type>
    <publisherName>Gesellschaft für Informatik</publisherName>
    <publisherPlace>Bonn</publisherPlace>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>1</belongsToBibliography>
    <completedDate>2024-12-12</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">Exploiting Data Redundancy in Im2Col Convolutions</title>
    <parentTitle language="deu">PARS-Mitteilungen 2025</parentTitle>
    <identifier type="issn">0177-0454</identifier>
    <identifier type="url">https://dl.gi.de/items/156c727e-ec90-4809-9ce6-ee27cc89eeb0</identifier>
    <enrichment key="THI_articleversion">published</enrichment>
    <enrichment key="THI_review">peer-review</enrichment>
    <enrichment key="THI_openaccess">nein</enrichment>
    <enrichment key="opus.source">publish</enrichment>
    <enrichment key="THI_conferenceName">30. PARS-Workshop, Ingolstadt (Germany), 25.-26.09.2024</enrichment>
    <author>
      <first_name>Tobias</first_name>
      <last_name>Pfaller</last_name>
    </author>
    <author>
      <first_name>Richard</first_name>
      <last_name>Membarth</last_name>
    </author>
    <collection role="institutes" number="19311">Fakultät Elektro- und Informationstechnik</collection>
    <collection role="institutes" number="19379">AImotion Bavaria</collection>
    <collection role="persons" number="43674">Membarth, Richard</collection>
  </doc>
  <doc>
    <id>1379</id>
    <completedYear/>
    <publishedYear>2019</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst>2164</pageFirst>
    <pageLast>2171</pageLast>
    <pageNumber/>
    <edition/>
    <issue/>
    <volume/>
    <articleNumber/>
    <type>conferenceobject</type>
    <publisherName>IEEE</publisherName>
    <publisherPlace>Piscataway</publisherPlace>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>1</belongsToBibliography>
    <completedDate>2022-02-22</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">Parallel Multi-Hypothesis Algorithm for Criticality Estimation in Traffic and Collision Avoidance</title>
    <parentTitle language="eng">2019 IEEE Intelligent Vehicles Symposium (IV)</parentTitle>
    <identifier type="isbn">978-1-7281-0560-4</identifier>
    <enrichment key="THI_review">peer-review</enrichment>
    <enrichment key="THI_openaccess">nein</enrichment>
    <enrichment key="THI_conferenceName">2019 IEEE Intelligent Vehicles Symposium (IV), Paris (France), 09.-12.06.2019</enrichment>
    <enrichment key="opus.source">publish</enrichment>
    <enrichment key="THI_relatedIdentifier">https://doi.org/10.1109/IVS.2019.8814015</enrichment>
    <enrichment key="THI_furtherversion">https://opus4.kobv.de/opus4-haw/frontdoor/index/index/docId/6714</enrichment>
    <author>
      <first_name>Eduardo</first_name>
      <last_name>Sánchez Morales</last_name>
    </author>
    <author>
      <first_name>Richard</first_name>
      <last_name>Membarth</last_name>
    </author>
    <author>
      <first_name>Andreas</first_name>
      <last_name>Gaull</last_name>
    </author>
    <author>
      <first_name>Philipp</first_name>
      <last_name>Slusallek</last_name>
    </author>
    <author>
      <first_name>Tobias</first_name>
      <last_name>Dirndorfer</last_name>
    </author>
    <author>
      <first_name>Alexander</first_name>
      <last_name>Kammenhuber</last_name>
    </author>
    <author>
      <first_name>Christoph</first_name>
      <last_name>Lauer</last_name>
    </author>
    <author>
      <first_name>Michael</first_name>
      <last_name>Botsch</last_name>
    </author>
    <collection role="institutes" number="15984">Fakultät Elektrotechnik und Informatik</collection>
    <collection role="institutes" number="15985">Fakultät Maschinenbau</collection>
    <collection role="persons" number="26835">Botsch, Michael</collection>
    <collection role="persons" number="27040">Gaull, Andreas</collection>
    <collection role="persons" number="43674">Membarth, Richard</collection>
    <collection role="institutes" number="19574">CARISSMA</collection>
  </doc>
</export-example>
