Forschungsbereich Parallel Computing

Organization Name (de) Name der Organisation (de)
E191-04 - Forschungsbereich Parallel Computing
 
Code Kennzahl
E191-04
 
Type of Organization Organisationstyp
Research Division
Parent OrgUnit Übergeordnete Organisation
 
Active Aktiv
 


Results 1-100 of 239 (Search time: 0.001 seconds).

PreviewAuthors / EditorsTitleTypeIssue Date
1Gao, Qi ; Han, Li ; Hunold, Sascha ; Robert, Yves ; Vivien, Frédéric Coping With Silent Errors for Workflows of Moldable TasksArticle Artikel 1-Nov-2026
2Hunold-2026-Parallel Computing-vor.pdf.jpgHunold, Sascha ; Träff, Jesper Larsson ; Laso, Ruben Tuned your MPI library? Now check the performance guidelinesArticle Artikel Sep-2026
3Laso, Ruben ; Salimi Beni, Majid ; Vardas, Ioannis ; Benkner, Siegfried ; Hunold, Sascha Where Did My NCCL Calls Go? A Profiler ComparisonInproceedings Konferenzbeitrag 19-Aug-2026
4Traeff-2026-Effective MPI User-defined Datatypes and Cartesian Communicato...-ao.pdf.jpgTräff, Jesper Larsson Effective MPI: User-defined Datatypes and Cartesian Communicators for Zero-copy All-to-all Communication in Multidimensional ToriPreprint Preprint 28-May-2026
5Traeff-2026-Two Efficient Message-passing Exclusive Scan Algorithms-ao.pdf.jpgTräff, Jesper Larsson Two Efficient Message-passing Exclusive Scan AlgorithmsPreprint Preprint 28-Apr-2026
6Laso-2026-To ncclsee, or Not to ncclsee That is the Profiling Question-vor.pdf.jpgLaso, Ruben ; Salimi Beni, Majid ; Vardas, Ioannis ; Benkner, Siegfried ; Hunold, Sascha To ncclsee, or Not to ncclsee: That is the Profiling QuestionInproceedings Konferenzbeitrag 10-Apr-2026
7Träff, Jesper Larsson Lectures on Parallel ComputingBook Buch2026
8Schuchart, Joseph ; Bouteiller, Aurelien ; Hunold, Sascha ; Jaeger, Julien ; Niethammer, Christoph ; Smith, Brian Recent Advances in the Message Passing InterfaceProceedings Tagungsband 2026
9Traeff-2025-ACM Transactions on Parallel Computing-vor.pdf.jpgTräff, Jesper Larsson Optimal, Non-pipelined Reduce-scatter and Allreduce Algorithms with an Application to All-to-all CommunicationArticle Artikel Dec-2025
10Pallez-2025-Implementing a Reproducibility Initiative in HPC Experiences ...-vor.pdf.jpgPallez, Guillaume ; Hill, Judith ; Hunold, Sascha Implementing a Reproducibility Initiative in HPC: Experiences from SC'24Inproceedings Konferenzbeitrag 21-Oct-2025
11Salimi Beni, Majid ; Laso, Ruben ; Cosenza, Biagio ; Benkner, Siegfried ; Hunold, Sascha Exploring NCCL Tuning Strategies for Distributed Deep LearningInproceedings Konferenzbeitrag 13-Aug-2025
12Vardas-2025-CONCURRENCY AND COMPUTATION-PRACTICE  EXPERIENCE-vor.pdf.jpgVardas, Ioannis ; Träff, Jesper Larsson ; Laso, Ruben ; Hunold, Sascha Mpisee: communicator-centric profiling of MPI applicationsArticle Artikel 25-Jul-2025
13Träff, Jesper Larsson Communication Round and Computation Efficient Exclusive Prefix-Sums Algorithms (for MPI_Exscan)Preprint Preprint 7-Jul-2025
14Vardas-2025-ncclsee A Lightweight Profiling Tool for NCCL-vor.pdf.jpgVardas, Ioannis ; Laso Rodriguez, Ruben ; Salimi Beni, Majid ncclsee: A Lightweight Profiling Tool for NCCLInproceedings Konferenzbeitrag 22-May-2025
15Salimi Beni-2025-Optimizing Distributed Deep Learning Training by Tuning ...-vor.pdf.jpgSalimi Beni, Majid ; Laso, Ruben ; Cosenza, Biagio ; Benkner, Siegfried ; Hunold, Sascha Optimizing Distributed Deep Learning Training by Tuning NCCLInproceedings Konferenzbeitrag 22-May-2025
16Hunold-2025-Report on Challenges of Practical Reproducibility for Systems ...-ao.pdf.jpgKeahey, Kate ; Richardson, Marc ; Tolosana Calasanz, Rafael ; Hunold, Sascha ; Lofstead, Jay ; Malik, Tanu ; Perez, Christian Report on Challenges of Practical Reproducibility for Systems and HPC Computer SciencePreprint Preprint 3-May-2025
17Traeff-2024-Optimal, Non-pipelined Reduce-scatter and Allreduce Algorithms-smur.pdf.jpgTräff, Jesper Larsson Optimal, Non-pipelined Reduce-scatter and Allreduce AlgorithmsPreprint Preprint 13-Feb-2025
18Carpentieri, Lorenzo ; De Caro, Antonio ; Salimibeni, Majid ; Fan, Kaijie ; Cosenza, Biagio Phase-Based Frequency Scaling for Energy-Efficient Heterogeneous ComputingInproceedings Konferenzbeitrag 2025
19Lin, Weiwei ; Jia, Zhen ; Hunold, Sascha ; Kang, Guoxin Benchmarking, Measuring, and Optimizing : 16th BenchCouncil International Symposium, Bench 2024, Guangzhou, China, December 4–6, 2024, Revised Selected PapersProceedings Tagungsband 2025
20Salimibeni, Majid ; Cosenza, Biagio ; Hunold, Sascha MPI Collective Algorithm Selection in the Presence of Process Arrival PatternsInproceedings Konferenzbeitrag 7-Nov-2024
21Vardas, Ioannis ; Hunold, Sascha ; Swartvagher, Philippe ; Träff, Jesper Larsson Exploring Mapping Strategies for Co-allocated HPC ApplicationsInproceedings Konferenzbeitrag 4-Nov-2024
22Traeff-2024-Lectures on Parallel Computing-smur.pdf.jpgTräff, Jesper Larsson Lectures on Parallel ComputingPreprint Preprint 26-Jul-2024
23Träff, Jesper Larsson Optimal Broadcast Schedules in Logarithmic Time with Applications to Broadcast, All-Broadcast, Reduction and All-ReductionPreprint Preprint 26-Jul-2024
24Laso Rodriguez-2024-pSTL-Bench A Micro-Benchmark Suite for Assessing Scal...-vor.pdf.jpgLaso Rodriguez, Ruben ; Krupitza, Diego ; Hunold, Sascha pSTL-Bench: A Micro-Benchmark Suite for Assessing Scalability of C++ Parallel STL ImplementationsPreprint Preprint 9-Feb-2024
25Salimi Beni-2024-Journal of Supercomputing-vor.pdf.jpgSalimi Beni, Majid ; Hunold, Sascha ; Cosenza, Biagio Analysis and prediction of performance variability in large-scale computing systemsArticle Artikel 2024
26Vardas, Ioannis ; Hunold, Sascha ; SWARTVAGHER, Philippe ; Träff, Jesper Larsson Improved Parallel Application Performance and Makespan by Colocation and Topology-aware Process MappingInproceedings Konferenzbeitrag 2024
27Hunold, Sascha ; Xie, Biwei ; Shu, Kai Benchmarking, Measuring, and Optimizing : 15th BenchCouncil International Symposium, Bench 2023, Revised Selected PapersProceedings Tagungsband2024
28Laso Rodriguez, Ruben ; Krupitza, Diego ; Hunold, Sascha Exploring Scalability in C++ Parallel STL ImplementationsInproceedings Konferenzbeitrag 2024
29Träff, Jesper Larsson ; Vardas, Ioannis ; Hunold, Sascha Modes, Persistence and Orthogonality: Blowing MPI UpInproceedings Konferenzbeitrag 2024
30Traeff-2023-Round-optimal n-Block Broadcast Schedules in Logarithmic Time-smur.pdf.jpgTräff, Jesper Larsson Round-optimal 𝑛-Block Broadcast Schedules in Logarithmic TimePreprint Preprint 18-Dec-2023
31Hunold, Sascha Unveiling the Complexities of Performance Analysis and Optimization in HPC SystemsPresentation Vortrag8-Dec-2023
32Swartvagher-2023-Using Mixed-Radix Decomposition to Enumerate Computation...-vor.pdf.jpgSwartvagher, Philippe ; Hunold, Sascha ; Träff, Jesper Larsson ; Vardas, Ioannis Using Mixed-Radix Decomposition to Enumerate Computational Resources of Deeply Hierarchical ArchitecturesInproceedings Konferenzbeitrag 12-Nov-2023
33Hunold-2023-Verifying Performance Guidelines for MPI Collectives at Scale-vor.pdf.jpgHunold, Sascha Verifying Performance Guidelines for MPI Collectives at ScaleInproceedings Konferenzbeitrag 12-Nov-2023
34Laso Rodriguez, Ruben ; Casado, Fernando E. The research career after the PhDPresentation Vortrag3-Nov-2023
35Traeff-2023-Library Development with MPI Attributes, Request Objects, Gro...-vor.pdf.jpgTräff, Jesper Larsson ; Vardas, Ioannis Library Development with MPI: Attributes, Request Objects, Group Communicator Creation, Local Reductions, and DatatypesInproceedings Konferenzbeitrag 21-Sep-2023
36Schuchart, Joseph ; Hunold, Sascha ; Bosilca, George Synchronizing MPI Processes in Space and TimeInproceedings KonferenzbeitragSep-2023
37Forsell, Martti ; Roivainen, Jussi ; Leppänen, Ville ; Träff, Jesper Larsson Realizing multioperations and multiprefixes in Thick Control Flow processorsArticle Artikel Apr-2023
38Träff, Jesper Larsson ; Hunold, Sascha ; Vardas, Ioannis ; Funk, Nikolaus Manes Uniform Algorithms for Reduce-scatter and (most) other Collectives for MPIInproceedings Konferenzbeitrag 2023
39Hunold, Sascha ; Steiner, Sebastian OMPICollTune: Autotuning MPI Collectives by Incremental Online LearningInproceedings Konferenzbeitrag 2023
40Hunold-2023-Massively Scaling Molecular Screening Workloads on EuroHPC Su...-vor.pdf.jpgHunold, Sascha ; Vardas, Ioannis ; Ibis, Gökhan ; Langer, Thierry Massively Scaling Molecular Screening Workloads on EuroHPC SupercomputersInproceedings Konferenzbeitrag 2023
41Vardas-2023-Effects of Mapping Strategies on Average Duration and Through...-vor.pdf.jpgVardas, Ioannis ; Hunold, Sascha ; Swartvagher, Philippe ; Träff, Jesper Larsson Effects of Mapping Strategies on Average Duration and Throughput of Colocated HPC ApplicationsInproceedings Konferenzbeitrag 2023
42Forsell, Martti ; Roivainen, Jussi ; Leppänen, Ville ; Träff, Jesper Larsson Preliminary Performance and Memory Access Scalability Study of Thick Control Flow ProcessorsInproceedings Konferenzbeitrag2023
43Hunold-2023-MPI is Good, Control is Better Checking Performance Guideline...-vor.pdf.jpgHunold, Sascha ; Hagn, Maximilian MPI is Good, Control is Better: Checking Performance Guidelines of CollectivesInproceedings Konferenzbeitrag 2023
44Swartvagher-2023-Rank Reordering within MPI Communicators to Exploit Deep...-vor.pdf.jpgSwartvagher, Philippe ; Vardas, Ioannis ; Hunold, Sascha ; Träff, Jesper Larsson Rank Reordering within MPI Communicators to Exploit Deep Hierarchal Architectures of SupercomputersInproceedings Konferenzbeitrag 2023
45Hunold, Sascha ; Kraßnitzer, Klaus Dieter Vincenz A Quantitative Analysis of OpenMP Task Runtime SystemsInproceedings Konferenzbeitrag2023
46Träff, Jesper Larsson Brief Announcement: Fast(er) Construction of Round-optimal n-Block Broadcast SchedulesInproceedings Konferenzbeitrag 11-Jul-2022
47Hunold-2022-An Overhead Analysis of MPI Profiling and Tracing Tools-vor.pdf.jpgHunold, Sascha ; Ajanohoun, Jordy Innocentius ; Vardas, Ioannis ; Träff, Jesper Larsson An Overhead Analysis of MPI Profiling and Tracing ToolsInproceedings Konferenzbeitrag 27-Jun-2022
48Hunold, Sascha ; Przybylski, Bartłomiej Scheduling.jl - Collaborative and Reproducible Scheduling Research with JuliaPresentation Vortrag18-May-2022
49Vardas, Ioannis ; Hunold, Sascha ; Ajanohoun, Jordy I. ; Traff, Jesper Larsson mpisee: MPI Profiling for Communication and Communicator StructureInproceedings Konferenzbeitrag 2022
50Hunold, Sascha Performance Tuning of MPI Collectives - Status Quo and Open ProblemsPresentation Vortrag2022
51Vardas, Ioannis ; Hunold, Sascha ; Ajanohoun, Jordy I. ; Träff, Jesper Larsson mpisee: MPI Profiling for Communication and Communicator StructureKonferenzbeitrag Inproceedings 2022
52Ajanohoun, Jordy I. ; Vardas, Ioannis ; Träff, Jesper Larsson ; Hunold, Sascha MPI Performance Tools under the Microscope: A Thorough Overhead AnalysisKonferenzbeitrag Inproceedings 2022
53Forsell, Martti ; Nikula, Sara ; Roivainen, Jussi ; Leppänen, Ville ; Träff, Jesper Larsson Performance and programmability comparison of the thick control flow architecture and current multicore processorsArtikel Article 2022
54Träff, Jesper Larsson Fast(er) Construction of Round-optimal n-Block Broadcast SchedulesInproceedings Konferenzbeitrag 2022
55Träff, Jesper Larsson (Poly)Logarithmic Time Construction of Round-optimal n-Block Broadcast Schedules for Broadcast and irregular Allgather in MPIPreprint Preprint2022
56Hunold, Sascha ; Przybylski, Bartlomiej Teaching Complex Scheduling AlgorithmsKonferenzbeitrag Inproceedings 2021
57Träff, Jesper Larsson ; Pöter, Manuel A more pragmatic implementation of the lock-free, ordered, linked listKonferenzbeitrag Inproceedings 2021
58Träff, Jesper Larsson ; Hunold, Sascha ; Mercier, Guillaume ; Holmes, Daniel J. MPI collective communication through a single set of interfaces: A case for orthogonalityArtikel Article 2021
59Hunold, Sascha ; Ajanohoun, Jordy I. ; Carpen-Amarie, Alexandra MicroBench Maker: Reproduce, Reuse, ImproveKonferenzbeitrag Inproceedings 2021
60Träff, Jesper Larsson A Doubly-pipelined, Dual-root Reduction-to-all Algorithm and ImplementationPreprint Preprint2021
61Träff, Jesper Larsson Decomposing MPI Collectives for Exploiting Multi-lane CommunicationPräsentation Presentation2020
62Kirchbach, Konrad Von ; Schulz, Christian ; Träff, Jesper Larsson Better Process Mapping and Sparse Quadratic AssignmentArtikel Article 2020
63Hunold, Sascha ; Bhatele, Abhinav ; Bosilca, George ; Knees, Peter Predicting MPI Collective Communication Performance Using Machine LearningKonferenzbeitrag Inproceedings 2020
64Träff, Jesper Larsson ; Hunold, Sascha Decomposing MPI Collectives for Exploiting Multi-lane CommunicationKonferenzbeitrag Inproceedings 2020
65Träff, Jesper Larsson Signature Datatypes for Type Correct Collective Operations, RevisitedKonferenzbeitrag Inproceedings 2020
66Faraj, Marcelo Fonseca ; van der Grinten, Alexander ; Meyerhenke, Henning ; Träff, Jesper Larsson ; Schulz, Christian High-Quality Hierarchical Process MappingKonferenzbeitrag Inproceedings 2020
67Hunold, Sascha ; Przybylski, Bartlomiej Scheduling.jl - Collaborative and Reproducible Scheduling Research with JuliaPreprint Preprint2020
68Forsell, Martti ; Roivainen, Jussi ; Träff, Jesper Larsson Optimizing Memory Access in TCF Processors with Compute-Update OperationsKonferenzbeitrag Inproceedings 2020
69von Kirchbach, Konrad ; Lehr, Markus ; Hunold, Sascha ; Schulz, Christian ; Träff, Jesper Larsson Efficient Process-to-Node Mapping Algorithms for Stencil ComputationsKonferenzbeitrag Inproceedings 2020
70Träff, Jesper Larsson ; Pöter, Manuel A more Pragmatic Implementation of the Lock-free, Ordered, Linked ListPreprint Preprint2020
71Faraj, Marcelo Fonseca ; van der Grinten, Alexander ; Meyerhenke, Henning ; Träff, Jesper Larsson ; Schulz, Christian High-Quality Hierarchical Process MappingPreprint Preprint2020
72Hunold, Sascha ; von Kirchbach, Konrad ; Lehr, Markus ; Schulz, Christian ; Träff, Jesper Larsson Efficient Process-to-Node Mapping Algorithms for Stencil ComputationsPreprint Preprint2020
73Träff, Jesper Larsson k-ported vs. k-lane Broadcast, Scatter, and Alltoall AlgorithmsPreprint Preprint2020
74Träff, Jesper Larsson ; Hoefler, Torsten Special issue: Selected papers from EuroMPI 2019Article Artikel2020
75Träff, Jesper Larsson Exploiting Multi-lane Communication in MPI CollectivesKonferenzbeitrag Inproceedings 2020
76Lehr, Markus ; von Kirchbach, Konrad Improved Cartesian Topology Mapping in MPIKonferenzbeitrag Inproceedings 2020
77Pachajoa, Carlos ; Levonyak, Markus ; Pacher, Christina ; Träff, Jesper Larsson ; Gansterer, Wilfried Classical and pipelined preconditioned conjugate gradient methods with node-failure resilienceKonferenzbeitrag Inproceedings 2020
78Träff, Jesper Larsson ; Hunold, Sascha ; Mercier, Guillaume ; Holmes, Daniel J. Collectives and Communicators: A Case for Orthogonality: (Or: How to get rid of MPI neighbor and enhance Cartesian collectives)Konferenzbeitrag Inproceedings 2020
79Träff, Jesper Larsson On Optimal Trees for Irregular Gather and Scatter CollectivesArtikel Article 1-Sep-2019
80Kang, Qiao ; Träff, Jesper Larsson ; Al-Bahrani, Reda ; Agrawal, Ankit ; Choudhary, Alok ; Liao, Wei-keng Scalable Algorithms for MPI Intergroup Allgather and AllgathervArtikel Article 2019
81Träff, Jesper Larsson On Optimal Trees for Irregular Gather and Scatter CollectivesPräsentation Presentation2019
82Träff, Jesper Larsson On optimal Trees for irregular gather and scatter collectives?Präsentation Presentation2019
83Träff, Jesper Larsson Cartesian Collective Communication: "Advice to users", "Advice to implementers", and "Advice to Standardizers"Präsentation Presentation2019
84Träff, Jesper Larsson On optimal Trees for irregular gather and scatter collectives?Präsentation Presentation2019
85Kainrad, Thomas ; Hunold, Sascha ; Seidel, Thomas ; Langer, Thierry LigandScout Remote: A New User-Friendly Interface for HPC and Cloud ResourcesArtikel Article 2019
86Hoefler, Torsten ; Träff, Jesper Larsson Proceedings of the 26th European MPI Users' Group Meeting, EuroMPI 2019Konferenzband Proceedings 2019
87Träff, Jesper Larsson ; Hunold, Sascha Cartesian Collective CommunicationKonferenzbeitrag Inproceedings 2019
88Kainer, Michael ; Träff, Jesper Larsson More Parallelism in Dijkstra's Single-Source Shortest Path AlgorithmPreprint Preprint2019
89Träff, Jesper Larsson Decomposing Collectives for Exploiting Multi-lane CommunicationPreprint Preprint2019
90Pachajoa, Carlos ; Levonyak, Markus ; Gansterer, Wilfried N. ; Träff, Jesper Larsson How to Make the Preconditioned Conjugate Gradient Method Resilient Against Multiple Node FailuresKonferenzbeitrag Inproceedings 2019
91Pachajoa, Carlos ; Levonyak, Markus ; Gansterer, Wilfried ; Träff, Jesper Larsson How to Make the Preconditioned Conjugate Gradient Method Resilient Against Multiple Node FailuresPreprint Preprint2019
92Träff, Jesper Larsson ; Hoefler, Torsten Foreword EuroMPI 2019Konferenzbeitrag Inproceedings2019
93Hunold, Sascha ; Carpen-Amarie, Alexandra On the Importance of Data Quality when Tuning MPI LibrariesKonferenzbeitrag Inproceedings 2019
94Forsell, Martti ; Roivainen, Jussi ; Leppänen, Ville ; Träff, Jesper Larsson Supporting concurrent memory access in TCF processor architecturesArtikel Article 2018
95Träff, Jesper Larsson Practical, distributed, low overhead algorithms for irregular gather and scatter collectivesArtikel Article 2018
96Pöter, Manuel ; Träff, Jesper Larsson <i>Stamp-it</i> , amortized constant-time memory reclamation in comparison to five other schemesKonferenzbeitrag Inproceedings 2018
97Pöter, Manuel ; Träff, Jesper Larsson Stamp-it: A more Thread-efficient, Concurrent Memory Reclamation Scheme in the C++ Memory ModelPreprint Preprint2018
98Träff, Jesper Larsson Parallel Quicksort without Pairwise Element ExchangePreprint Preprint2018
99Pöter, Manuel ; Träff, Jesper Larsson Memory Models for C/C++ ProgrammersPreprint Preprint2018
100Hunold, Sascha ; Carpen-Amarie, Alexandra Hierarchical Clock Synchronization in MPIKonferenzbeitrag Inproceedings 2018