HPC System Engineer

Role overview

Infrastructure and scheduling

• Maintain compute nodes, login nodes, heterogeneous nodes, NAS storage servers, and InfiniBand interconnects.
• Configure and tune Slurm, PBS, LSF, or Grid Engine schedulers.
• Implement fair-share scheduling, job priorities, QoS limits, reservations, and preemption policies.
• Recommend CPU and GPU architectures for engineering simulations.

Security and software

• Manage accounts, permissions, storage quotas, secure SSH access, MFA, patches, and vulnerability fixes.
• Maintain compliance with IT governance, data-security, and audit requirements.
• Install and manage ANSYS, Siemens, NASTRAN, environment modules, compilers, MPI libraries, and drivers.

Performance and support

• Monitor utilization, queue wait times, job failures, hardware health, and cluster performance.
• Resolve network congestion, I/O bottlenecks, memory pressure, job crashes, solver errors, and environment issues.
• Support batch scripting and job optimization, and provide user training and documentation.

Storage and capacity planning

• Enforce quotas, purge policies, retention rules, backups, archival systems, and RAID configurations.
• Maintain Lustre, GPFS, or BeeGFS parallel file systems for large CFD/FEA datasets.
• Plan expansion for additional cores, GPUs, memory-heavy nodes, and faster interconnects.

Basic qualifications

• Bachelor's or Master's degree in Computer Science, Engineering, Physics, or a related field.
• 3–7 years of experience administering Linux-based HPC clusters.
• Strong knowledge of HPC schedulers such as Slurm, PBS, or LSF.
• Experience with MPI, OpenMP, and distributed computing.
• Familiarity with CFD/FEA solvers and engineering workflows.
• Proficiency in Bash or Python scripting.
• Experience with high-speed networking and parallel file systems.

Preferred qualifications

• Experience with GPU-accelerated workloads and CUDA-enabled solvers.
• Knowledge of Singularity or Apptainer container technologies.
• Experience with Grafana, Prometheus, or XDMoD monitoring tools.
• Background supporting engineering or scientific computing environments.
• Understanding of performance profiling and benchmarking.

Additional role requirement

• English proficiency is required for all roles unless otherwise stated in the posting.