Stijn Heldens🔊

Curriculum Vitae

2022 - present

Netherlands eScience Center. Research Software Engineer. Focus on accelerating scientific projects using HPC and GPU programming.

2022 - present

University of Amsterdam Visiting researcher.

2018 - 2022

Netherlands eScience Center / University of Amsterdam PhD on the topic of programming models towards exascale computing.

2016 - 2017

University of Twente. Researcher at Stochastic Operations Research Group. Worked on processing of large-scale spatio-temporal graphs.

2015 - 2016

Delft University of Technology. Researcher at Distributed Systems group. Worked on graph processing using heterogeneous CPU-GPU systems.

2012 - 2015

VU University Amsterdam. Master Computer Science, specialization tracks High Performance Distributed Computing and Internet Programming.

2009 - 2012

University of Amsterdam Bachelor Computer Science.

Publications

• M Lurati, S Heldens, A Scloccco, B van Werkhoven Bringing Auto-tuning to HIP: Analysis of Tuning Impact and Difficulty on AMD and Nvidia GPUs International European Conference on Parallel and Distributed Computing (Euro-Par) 2024 10.1007/978-3-031-69577-3_7
  • preprint
  • paper
  • presentation
• S Heldens Parallel Programming Systems for Scalable Scientific Computing PhD dissertation, University of Amsterdam 2024
  • thesis
• P Hijma, S Heldens, A Sclocco, B van Werkhoven, H Bal Optimization Techniques for GPU Programming ACM Computing Surveys 2023 10.1145/3570638
  • paper
• S Heldens, B van Werkhoven Kernel Launcher: C++ Library for optimal performance-portable CUDA kernels IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW) 2023 10.1109/IPDPSW59300.2023.00126
  • preprint
  • paper
  • code
• S Heldens, A Sclocco, H Dreuning, B van Werkhoven, P Hijma, J Maassen, RV van Nieuwpoort litstudy: A Python package for literature reviews SoftwareX 2022 10.1016/j.softx.2022.101207
  • paper
  • code
• S Heldens, P Hijma, B van Werkhoven, J Maassen, RV van Nieuwpoort Lightning: Scaling the GPU Programming Model Beyond a Single GPU IEEE International Parallel and Distributed Processing Symposium (IPDPS) 2022 10.1109/IPDPS53621.2022.00054
  • preprint
  • paper
  • code
• S Heldens, P Hijma, B van Werkhoven, J Maassen, RV van Nieuwpoort Rocket: Efficient and Scalable All-Pairs Computations on Heterogeneous Platforms Conference for High Performance Computing, Networking, Storage, and Analysis (SC20) 2020 10.1109/SC41405.2020.00105
  • preprint
  • paper
  • presentation
  • code
• S Heldens, P Hijma, B van Werkhoven, J Maassen, ASZ Belloum, RV van Nieuwpoort The Landscape of Exascale Research - Data-Driven Literature Analysis ACM Computing Surveys 2020 10.1145/3372390
  • paper (open access)
• A Iosup, T Hegeman, WL Ngai, S Heldens, et al. LDBC Graphalytics: A benchmark for large-scale graph analysis on parallel and distributed platforms PVLDB 2016 10.14778/3007263.3007270
  • preprint
  • paper
• S Heldens, AL Varbanescu, A Iosup Dynamic load balancing for high-performance graph processing on hybrid CPU-GPU platforms Workshop on Irregular Applications: Architectures and Algorithms 2016 10.1109/IA3.2016.016
  • paper
  • code
• S Heldens, A Varbanescu, WL Ngai, T Hegeman, A Iosup Towards the Next Generation of Large-Scale Network Archives European Conference on Parallel Processing 2016 10.1007/978-3-319-58943-5_46
  • paper
• S Heldens, AL Varbanescu, A Prat-Pérez, JL Larriba-Pey Towards community detection on heterogeneous platforms European Conference on Parallel Processing 2015 10.1007/978-3-319-27308-2_18
  • paper
  • code

Related Software

• Kernel Tuner Ecosystem: Software for the creation of highly-optimized GPU applications through auto-tuning.
• Kernel Launcher: Launching CUDA kernels which have been tuned by Kernel Tuner.
• Kernel Float: header-only library for reduced precision in CUDA.
• Lightning: Data-oriented lightning-fast parallel processing using GPUs on distributed clusters.
• HyGraph: High-performance graph processing on hybrid CPU-GPU platforms by using dynamic load-balancing.
• Par-CD: Flexible Framework for Parallel Community Detection in Large Networks on multi-core CPUs and GPU.

Contact

• Google Scholar: Stijn Heldens
• Github: stijn.h
• LinkedIn: Stijn Heldens
• DBLP: Stijn Heldens