Senior HPC Systems Administrator

About the role

We are seeking a highly motivated and skilled Senior HPC Systems Administrator to join the KAUST Supercomputing Laboratory. The role involves managing an HPC cluster of approximately 600 CPU and GPU nodes, storage systems, InfiniBand and Ethernet networks, and providing broad support to researchers across computational science, engineering, big data analysis, and AI/ML workloads.

Key responsibilities

• Provide timely user support via phone, walk‑in, email and ticketing system while maintaining high customer‑service standards.
• Install, configure and manage HPC subsystems including compute nodes, high‑performance storage, InfiniBand, Ethernet and configuration‑management tools such as Ansible or Puppet.
• Deploy and maintain cluster management software, monitoring tools and supporting services for the HPC environment.
• Administer the Slurm workload manager, manage QoS policies, accounts, accounting and develop automation scripts (Python, C++).
• Develop and maintain Bash and Python automation scripts to streamline system‑administration tasks.
• Deploy and manage container environments (Singularity/Apptainer, Docker) for HPC workloads.
• Benchmark CPU, memory, InfiniBand and storage components to ensure optimal performance and identify tuning opportunities.
• Enforce security best practices, including node hardening, kernel patching and compliance across all systems.

Required profile

• Strong expertise in Linux system administration (RHEL, Rocky Linux or CentOS) within large‑scale HPC environments.
• Demonstrated track record managing complex HPC systems, including parallel file systems, job schedulers, high‑speed networks and monitoring solutions.
• Experience supporting users of computational science, engineering, data analysis and AI/ML applications.
• Ability to collaborate with researchers, application teams and vendors in a multicultural, international setting.
• Excellent written and verbal communication skills in English.

Required skills

• Linux (RHEL, Rocky Linux, CentOS)
• Slurm workload manager
• Ansible or Puppet
• Bash scripting
• Python scripting
• C++ programming
• Singularity/Apptainer and Docker containers
• InfiniBand and Ethernet networking
• Lustre, GPFS, Weka or Vast parallel file systems
• MPI, OpenMP, CUDA, OpenACC
• Fortran, C/C++