Ben van Werkhoven is assistant professor at LIACS and head of the Accelerated Computing research group. He is also affiliated with the Netherlands eScience Center. His research interests lie in High Performance Computing (HPC), software optimization, automatic performance tuning (auto-tuning), energy efficiency, programming models, performance modeling, and the acceleration of scientific applications.

More information about Ben van Werkhoven

Ben van Werkhoven is assistant professor at LIACS and head of the Accelerated Computing research group. He is also affiliated with the Netherlands eScience Center. His research interests lie in High Performance Computing (HPC), software optimization, automatic performance tuning (auto-tuning), energy efficiency, programming models, performance modeling, and the acceleration of scientific applications.

Ben van Werkhoven is leading the Accelerated Computing research group. This group conducts research in accelerating scientific or compute-intensive applications for performance, energy efficiency, and accuracy using automated techniques, including optimization algorithms and machine learning techniques, including large language models, for advanced and emerging hardware platforms, including supercomputers and Graphics Processing Units (GPUs).

His recent research achievements include high-impact publications at the absolute top venues in computer science. Including surveys on Exascale computing and Optimization techniques for GPU programming, both published in ACM Computing Surveys, programming models for highly-optimized distributed GPU applications, including Rocket published at Supercomputing (SC), and Lightning published at International Parallel and Distributed Processing Symposium (IPDPS).

His most successful project is Kernel Tuner, which has featured in tutorials and presentations at world-leading science and industry events including Nvidia’s GPU Technology Conference (GTC), ACM/IEEE Supercomputing Conference (SC), and International Supercomputing (ISC-HPC), and has been used to, among others, enable new scientific methods in localization microscopy, published in Nature Methods and Nature Communications.

Teaching

Ben van Werkhoven is passionate about teaching and has created two new courses since he started at LIACS. In the Master Computer Science, Ben is teaching his brand new High Performance Computing course. Starting in academic year 2024/2025, he is teaching the new Multiprocessor Programming course, a third year elective course in the Bachelor Computer Science.

Active projects

• Kernel Tuner
  Main developers: Ben van Werkhoven (creator), Alessio Sclocco, Stijn Heldens, Floris-Jan Willemsen, Willem Jan Palenstijn, Richard Schoonhoven, and Bram Veenboer
  
  Kernel Tuner is a software development tool for the creation of highly-optimized and tuned GPU applications. Ben van Werkhoven created Kernel Tuner in 2016, while working on several GPU applications. Recently, the project has grown from a single tool to an ecosystem of tools being built on top of or around Kernel Tuner. 
  Funding: various projects have contributed to the development of Kernel Tuner since 2016.

• CORTEX 
  The National Science Agenda (NWO NWA) has awarded a 5 million euro grant to CORTEX – the Center for Optimal, Real-Time Machine Studies of the Explosive Universe. The CORTEX consortium of 12 partners from academia, industry and society will make self-learning machines faster, to figure out how massive cosmic explosions work, and to innovate wider applications. Ben van Werkhoven is leading one of the work packages within CORTEX, where he investigates how to create software with the help of machine learning that can make optimal use of the computing power of modern computers. This technology is also applied within the work package to implement the software pipelines for observing explosive events in the universe. As part of CORTEX, we have developed new methods for model-steered auto-tuning to optimize the energy efficiency of several Radio Astronomy applications, click here for the paper.
   

• ESiWACE3
  The ESiWACE3 project (6 million euro grant funded by EuroHPC JU and national co-funders, including the Netherlands Enterprise Agency (RVO)) aims to advance Europe’s Earth system modeling capabilities. ESiWACE3 has the goal to promote efficient and scalable simulation of weather and climate, to close common technology knowledge gaps and provide toolboxes for high-performance computing for weather and climate modeling in Europe. Ben is leading work package 4 on HPC services, which creates collaborative projects to advance the use new architectures such as GPUs in weather and climate models. As part of work package 2, Ben is also working on extending the capabilities of Kernel Tuner to automatically optimize weather and climate models for high performance, energy efficiency and accuracy.

Past projects

• ConFu
  The Consolidating and Future-proofing Kernel Tuner by developing Software Engineering Best Practices (ConFu) project (Netherlands eScience Center, 40 K euro) had as its goal to consolidate the Kernel Tuner ecosystem and carry out several software engineering improvements, streamlining the software development processes, and prepare Kernel Tuner for new use cases in future projects and collaborations. The upgrades took place in several software layers and core of Kernel Tuner, as well as in the CI/CD infrastructure. Ben van Werkhoven is the PI of Kernel Tuner, and was the PI for the ConFu project.
   

• ESiWACE2
  The EU H2020 (8 M euro) funded Centre of Excellence in Simulation of Weather and Climate in Europe (ESiWACE) enables global storm- and eddy resolving weather and climate simulations on the upcoming (pre-)Exascale supercomputers. Ben van Werkhoven was leading the work package on HPC services, which created small collaborative projects that provide guidance, engineering, and advice to developers of weather and climate models. The aim was to improve model efficiency and to port models to new architectures such as GPUs. As part of ESiWACE2, Kernel Tuner was used to optimize and auto-tune several weather and climate simulating codes, including RTE-RRTMGP and MicroHH.
   

• eWaterCycle II
  Ben was co-PI of the eWaterCycle II (Netherlands eScience Center 1.3 M euro) project. The goal of eWaterCycle II was to develop a framework in which hydrological modellers can work together in a collaborative environment. This environment allows to easily combine models and data independent of programming languages, compare models to other models of the same area, and reproduce the results obtained by their peers.

Publications