HPC Systems Splst 2

The HPC Systems Specialist 2 provides systems administration, integration, deployment, and troubleshooting support for a large-scale high-performance computing environment. The position has two closely coupled technical areas of emphasis: (1) high-performance storage and data infrastructure and (2) the high-speed network fabrics that connect storage, compute and supporting systems.

If hired the candidate will support production HPC parallel filesystems such as IBM Spectrum Scale (GPFS), BeeGFS, Lustre, or comparable technologies, together with data-management services and data-intensive computing platforms that use or expose that storage, such as iRODS, CyVerse, Kubernetes, or comparable systems. The role also manages and troubleshoots High performance Ethernet and InfiniBand fabrics. Across both areas, the role requires strong Linux and production HPC systems knowledge and the ability to diagnose complex problems that cross storage, filesystem, network, hardware, firmware, operating-system, and application boundaries in a campus scale heterogenous network.

Bachelor’s degree; at least 3 years of experience directly related to the duties and responsibilities specified.

Completed degree(s) from an accredited institution that are above the minimum education requirement may be substituted for experience on a year for year basis.

Preferred Qualifications

The strongest candidates will have experience across the areas below. Candidates are not expected to have experience with every technology listed.

HPC Storage and Data Infrastructure

• Demonstrated hands-on experience deploying, administering, troubleshooting, or performance-tuning a production parallel filesystem in an HPC or other large-scale computing environment, such as IBM Spectrum Scale (GPFS), BeeGFS, Lustre, or a comparable HPC parallel filesystem. Direct administration of the parallel filesystem is strongly preferred; enterprise NAS/SAN administration alone is not considered equivalent.

• Experience integrating HPC storage with data-management or data-intensive computing platforms such as iRODS, CyVerse, Kubernetes, or comparable technologies. Experience with data movement, metadata, replication, data lifecycle management, object/data services, or storage integration for containerized workloads is desirable.

• Experience with HPC storage hardware and architectures, including storage servers, disk and flash storage, RAID or erasure-protected systems, storage enclosures, and high-bandwidth storage networks. High-Speed Networking and Fabrics

• Experience configuring, administering, monitoring, or troubleshooting InfiniBand switches and fabrics, particularly

NVIDIA/Mellanox technologies or comparable high-performance interconnects.

• Experience configuring, administering, monitoring, or troubleshooting enterprise or high-performance Ethernet switches and networks.
• Experience with high-performance networking concepts such as RDMA, fabric management, VLANs, bonding/link aggregation, routing, congestion/performance analysis, or network monitoring, especially in large heterogeneous environments.

HPC Systems Administration and Operations

• Strong hands-on Linux systems administration experience in production HPC or other large-scale computing environments, including cluster systems and services, Slurm or comparable batch scheduling, system monitoring, and utilization/performance analysis.
• Demonstrated ability to troubleshoot performance or reliability problems across subsystem boundaries, including storage, filesystems, networking, Linux, hardware, firmware, and applications.
• Experience with Bash, Python, Ansible, or comparable tools for automation, deployment, monitoring, and configuration management.
• Experience deploying, upgrading, documenting, securing, and maintaining production infrastructure, including coordination of changes across operating systems, drivers, firmware, storage/data software, networks, security controls, and vendor-supported hardware.

Additional Requirements Position requires: a) Using independent judgment on tasks and problem solving of a complex nature; b) working as member of Systems team and with project leader on sizeable projects; c) working as part of Systems team to deploy and configure large parallel and system software components; d) demonstrated depth of understanding of specialty area; e) working as member of Systems team to develop new ideas and guide the organization in technology; f) serving as mentor to other HPC systems engineers and technical staff.