HPC Systems Administrator at Santa Clara University in Santa Clara, CA

Job description

Position Title: HPC Systems Administrator Position Type: Regular Hiring Range: $129,000.00 - $161,265.00 annually; Compensation will be based on education, experience, skills relevant to the role, and internal equity. Pay Frequency: Annual A. POSITION PURPOSE The High-Performance Computing (HPC) System Administrator is a hands-on role responsible for the day-to-day configuration, monitoring, optimization, and operation of the organization's high-performance computing infrastructure. This individual will focus on system optimization, troubleshooting, and general planning for future infrastructure enhancements across compute, storage, and high-speed interconnects (InfiniBand). A key responsibility is to collaborate with existing system administrators, building the team’s collective HPC expertise, strengthening shared support capabilities, and ensuring long-term operational resilience and efficiency. The HPC Systems Administrator is a member of the Enterprise Systems team within the Cyberinfrastructure Technolog i es department. The incumbent works with the other Cyberinfrastructure teams - Network and Telecommunications, Enterprise Applications, and the Information Security Office - and other campus divisions in coordinating services, providing support and providing appropriate guidance. This incumbent will also work with University vendors and partners. The HPC Systems Administrator will have a passion for providing excellent customer serv i ce, and a focus on continual improvement; a commitment to supporting innovative infrastructure technologies; and a desire to learn and grow their technical expertise to meet the needs of the campus community. B. ESSENTIAL DUTIES AND RESPONSIBILITIES 1. HPC Infrastructure Management and Optimization Compute: Supports the lifecycle of all compute nodes, including installing, configuring, and maintaining hardware, operating systems, and core system software to ensure optimal performance, stability, and resource utilization for scientific workloads. Storage: Assists in the management of the high-performance parallel file systems (e.g., Lustre, GPFS, CephFS), NAS, and backup solutions, executing capacity planning, performance tuning, and integrity checks to guarantee secure, high-speed, and reliable data access for all users. InfiniBand: Designs, deploys, and provides troubleshooting and maintenance for the InfiniBand high-speed interconnect fabric, ensuring low-latency, high-bandwidth inter-node communication essential for scalable HPC application performance. 2. Workload Management and System Deployment Slurm: Supports day-to-day administration and configuration of the Slurm Workload Manager, monitoring job queues, partitions, and resource allocation policies to enforce fair-share scheduling, maximize cluster utilization, and meet diverse research computational needs. System Imaging: Helps develop, maintain, and update standardized, optimized system images for all compute nodes, utilizing automation tools to facilitate rapid, consistent deployment, efficient patching, and streamlined upgrades across the cluster environment. Software Licenses: Supports administration of all commercial scientific software licenses, helping track compliance with vendor agreements and managing license servers and usage policies to optimize utilization and accessibility for the HPC user base. 3. Professional Development and Knowledge Building Actively participates in cross-training and knowledge-sharing sessions for existing system administrators to enhance the team's collective expertise in HPC-specific technologies (Slurm, InfiniBand, parallel file systems). Documents routine procedures and troubleshooting steps to support team knowledge sharing and consistency. Stays current on emerging HPC technologies and relevant industry trends, evaluating vendor solutions, and providing recommendations with the team. 4. Coordination and Collaboration Participates and plays an active role in the planning and implementation p