HPC Advisor

Do you want to contribute to supporting research on the Dutch National Supercomputer Snellius and the LUMI pre-exascale Supercomputer? Are you familiar with scientific workflows and applications, experienced in performance analysis and optimisation and analysing CPU- and GPU-based workloads to support Dutch research and education? Then we have the perfect challenge for you!

Where you will work

SURF is the ICT cooperative for Dutch educational and research institutions. Together with them, we work on digital services and complex innovation challenges to enhance the quality of education and research. As a High Performance Computing (HPC) advisor, you will work with colleagues and international partners, contributing to strengthening the national and European HPC environment. You will also collaborate with other teams at SURF and contribute to understanding scientific applications and workflows and improving their performance,

The team you will join

You will be joining the HPC Services team. This team is responsible for hosting, maintaining, and supporting users on the National Supercomputer Snellius, LUMI, and in the future Alice Recoque. You will collaborate with the system administrators and other advisors in the HPC Services team but also work closely with the other HPC team at SURF – the HPC Development team.

Together the HPC Services and HPC Development teams are responsible for various computing services and the associated user and application support, including supporting the National Supercomputer Snellius, the LUMI pre-exascale Supercomputer, and SURF’s Experimental Technologies Platform. The teams focus, among other things, on providing users of the SURF (and international) infrastructures with permanent access to the most appropriate hardware and software solutions. We assist users in efficiently using new technologies and methods, enabling them to make effective use of the available infrastructure. In this way, we contribute to our mission of facilitating top-level Dutch research.

Next to the HPC teams at SURF, you will also join the LUMI User Support Team. This team is a diverse group of HPC advisors that are associated with HPC institutes in each of the eleven LUMI consortium countries. Together, they support users, provide training, and maintain LUMI’s.

What you will do

With your knowledge of scientific computing applications and programming models, you support researchers using state-of-the-art infrastructure with a focus on LUMI and Snellius. You also work on developing tools that improve system efficiency and user experience, drawing on modern HPC technologies (CPU and GPU architectures, accelerated computing) and methodologies (parallelization or application of artificial intelligence techniques).

Your tasks will include at least:

• Supporting users of the LUMI pre-exascale Supercomputer, including software stack dvelopment, testing, documentation, and best practices for Dutch and European research;
• Supporting users of the Dutch National Supercomputer, Snellius, and provides support to researchers who use the systems, including software stack development, testing, documentation, and best practices;
• Perform technical reviews for applications for computing time on LUMI and Snellius.
• Contributing to outreach activities such as developing course materials or delivering courses, participating in webinars, disseminating scientific and technical results;
• Staying up to date with relevant HPC developments.

Your skills and experience

As an HPC advisor, you have the ability to apply your in-depth knowledge of computer systems, and your hands-on experience in solving practical problems. You are analytically strong and have a proactive attitude.

In addition, you have:

• An academic thinking and working level: (MSc/PhD level) in Computer Science, Mathematics or other traditional HPC fields such as Chemistry, Physics or Astronomy;
• At least 2 years of relevant work experience;
• Experience with programming in compiled/interpreted languages such as Fortran, C/C++ Python, R, bash, as well as experience with parallel programming (MPI/OpenMP);
• Experience with GPU programming, knowledge of AMD GPUs is a plus (e.g., AMD ROCm/HIP, OpenCL);
• Experience with running scientific applications in a high-performance computing environment and good understanding of performance analysis/optimization tools and concepts;
• Good command of the English language in writing and speaking.

Prior to starting this job, a VOG must be presented.

SURF takes pleasure in doing its recruitment itself; acquisition is therefore not appreciated.