HPC Systems Engineer – On-Call & High-Perf Networking

University Corporation for Atmospheric Research; Boulder (CO); On-site; USD 68,000 - 85,000; Full time

Job summary

University Corporation for Atmospheric Research (UCAR) seeks an HPC Systems Engineer to support CISL high-performance computing, storage, and networks. The role involves engineering, installation, and ongoing maintenance of multi-vendor systems, with occasional travel to the NWSC in Cheyenne, WY.

The position requires Linux expertise, scripting, and programming, plus experience with HPC scheduling (PBS/SLURM), clustered file systems (Spectrum Scale, Lustre), and HPC storage.

Qualifications

• Experience with Linux OS, scripting, and programming languages.
• HPC and HPS software analysis, design, and development.
• Familiarity with HPC networking, high-performance interconnects, and security.
• Experience with installation and integration of HPC/HPS hardware and software.
• Ability to travel to NWSC near Boulder (Cheyenne, WY).
• Willingness for periodic 24x7 on-call duty.
• Experience with PBS and SLURM scheduling systems.
• Experience with clustered file systems such as Spectrum Scale and Lustre.
• Knowledge of HPC storage types and disaster recovery of clusters.

Responsibilities

• Install, configure, and troubleshoot Linux-based HPC systems and storage.
• Analyze, design, and develop HPC/HPS software and interconnects.
• Install/integrate HPC hardware/software; ensure security and reliability.
• Provide 24x7 on-call support and coordinate with CISL and vendors.

Skills

Linux

Scripting

Programming

HPC systems

Networking

Storage

On-call

Travel

Tools

PBS

SLURM

Spectrum Scale

Lustre

Git

Job description

University Corporation for Atmospheric Research (UCAR) seeks an HPC Systems Engineer to support CISL high-performance computing, storage, and networks. The role involves engineering, installation, and ongoing maintenance of multi-vendor systems, with occasional travel to the NWSC in Cheyenne, WY.

The position requires Linux expertise, scripting, and programming, plus experience with HPC scheduling (PBS/SLURM), clustered file systems (Spectrum Scale, Lustre), and HPC storage.