Senior HPC Systems Administrator, SAS Computing

University of Pennsylvania

3d•Hybrid

About The Position

The Linux Infrastructure Services (LIS) group at the School of Arts and Sciences (SAS) is seeking a passionate and skilled Senior HPC Systems Administrator. Join our team and collaborate with world-renowned researchers tackling questions about the human brain, the upper atmosphere, ocean biogeochemistry, social program impacts, and more. Under the guidance of the HPC team leadership, you will ensure the smooth operation of our HPC research services using technologies like GPU’s, CPU’s, Slurm, conda, infiniband, and many other components of a modern, shared HPC environment. This position is hybrid-eligible. The School of Arts and Sciences is the heart of Penn’s academic enterprise, home to hundreds of faculty and thousands of students engaged in innovative research and teaching across the liberal arts and sciences. SAS Computing provides comprehensive IT support for Penn’s world-class research, teaching and learning, and administrative activities. Our team works closely with faculty and staff to deliver innovative, forward-looking technology solutions that meet a wide range of needs. With a dedicated group of professionals known for strong collaboration and teamwork, SAS Computing translates technical expertise into practical support that advances the academic mission of the School.

Requirements

Bachelor of Science and 3 to 5 years of experience or equivalent combination of education and experience is required.
Proficiency in Linux OSes (Ubuntu preferred)
Advanced Linux scripting skills (BASH, Python, etc.)
A working knowledge of job scheduling systems (SLURM preferred)
Expertise in managing high-performance computing resources
Networking experience including Infiniband and TCP networks, and configuring and managing network architecture within an HPC environment
Proficiency in managing storage solutions and backups
Experience in using configuration management tools such as Salt, Ansible, and MAAS.
Experience in working with git repositories
Experience in deploying and managing HPC hardware systems, including CPU and GPU nodes, networks, and storage.
Skilled in triaging complex problems and developing solutions
Strong communication skills to maintain effective interactions with stakeholders and team members

Nice To Haves

Experience deploying and managing cloud-based linux resources (AWS preferred)
Ability to work collaboratively with SAS Computing colleagues, faculty, research staff, and other stakeholders
Capable of managing and tracking multiple ongoing projects simultaneously
Committed to the research and academic mission of SAS

Responsibilities

Serve as a Sr. Systems Administrator managing complex Linux systems.
Support research computing clusters, databases, and web servers.
Support SAS researchers who use Penn’s new PARCC (Penn Advanced Research Computing Center) centralized HPC services, including both CPU and GPU cutting-edge technologies.
Maintain, configure, and provide automation support for high-performance computing solutions in datacenters and the cloud.
Engage with researchers to understand how HPC can enhance and transform their work.
Proactively pursue efficient and collaborative solutions to requests, partnering with faculty and local computing support providers across the school.
Deploy and manage Linux systems
Develop shell and python scripts
Configure, manage, and optimize job scheduling software (SLURM)
Install and configure free and licensed software
Monitor systems and services
Perform routine systems maintenance
Manage data and configuration backups
Coordinate hardware repairs
Oversee ordering and installation of hardware
Recommend and track software and hardware changes
Automate systems configuration tasks and deployments
Provide technical consulting and end-user Linux support
Support web services
Assist first-tier support staff with end-users issues on our systems
Maintain expert-level knowledge of HPC technologies
Propose and implement improvements to our HPC services
Provide support for SAS researchers using Penn’s PARCC central HPC cluster and partner with the PARCC team to meet the needs of SAS faculty
Participate in the linux systems administration on-call rotations, which includes on-prem and cloud (AWS) services.