Senior HPC Systems Administrator (HR Title: Systems Administrator III)

About the position

This role is an on-campus, in-person position. Dedicated to supporting SMU's research community, the Senior System Administrator for High Performance Computing (HPC) works exclusively to design, build, maintain, operate and manage HPC systems at SMU. This position shares responsibility for university HPC technical support as member of a two-person HPC systems infrastructure team. This position also assists with Enterprise Linux support. This position provides hardware, software and end-user support for SMU's growing number of research faculty and center compute resources dedicated to advancement of SMU research activities. Demonstrates advanced knowledge with all the technical tools required to perform the job. Subject matter expert in primary areas of support. Able to solve complex problems crossing multiple research disciplines with little or no escalation support. Effective technical resource to others to resolve problems and implement projects.

Responsibilities

• Design, plan, deploy, administer services & troubleshoot issues related to HPC services for research at SMU.
• Install and maintain cluster environments and provision systems using automated installation methods.
• Manage/maintain Lustre parallel file system and NFS storage.
• Manage/maintain InfiniBand high performance interconnect fabric.
• Configure, manage, monitor SLURM scheduling & queuing system.
• Develop/maintain programs/scripts that aid in operation and automation of administrative tasks using various shell and scripting languages (bash, Perl, Python).
• Compile, install, and port software in support needed by SMU researchers.
• Build and deploy open source and vendor/commercial software required by researchers.
• Plan projects, communicate with end users and management, provide updates and expectations management.
• Document all configurations, procedures, and changes.
• Diagnose and resolve system and operational problems with research systems.
• Work with researchers and constituents to diagnose and optimize workloads.
• Participate in on call support of research infrastructure.
• Coordinate with vendors to resolve hardware and software problems.
• Ensure hardware firmware and software revision levels are maintained at the appropriate level on HPC research systems.
• Keep current with research computing, HPC technology trends and best practices.

Requirements

• Bachelor's degree is required.
• A minimum of six years of full time Linux system administration experience in a large computing environment is required.
• Candidate must demonstrate clear, professional communication to work with team members and customers of diverse technical abilities.
• Candidate must demonstrate strong written communication skills.
• Candidate must possess strong problem-solving skills with the ability to identify and analyze problems, as well as devise solutions.
• Candidate must have strong organizational, planning and time management skills.

Nice-to-haves

• Experience with NVidia DGX, Containers and Kubernetes.
• Knowledge of reporting tools including XDMO.
• Work experience installing and maintaining clustered environments and provisioning systems using automated installation methods.
• Direct experience working with InfiniBand and knowledge of configuration and management of SLURM or other scheduling and queuing systems.
• Familiarity with DDN hardware and the Lustre file system.
• Proficiency in supporting Nvidia/Mellanox InfiniBand networks.
• Competence with Bright Cluster Manager.
• Knowledge of Nvidia DGX systems.

Benefits

• Broad, competitive array of health and related benefits.
• Traditional benefits such as health, dental, and vision plans.
• Wide range of wellness programs.
• Array of retirement programs.
• Access to a wide variety of professional and personal development opportunities, including tuition benefits.