Senior Director, Advanced Systems (HPC)

Princeton University•Princeton, NJ

53d

About The Position

The Senior Director of Advanced Systems in the Research Computing department within the Office of the Dean for Research, leads a team of technical experts and system administrators that design, deploy, maintain, and support Princeton's centralized high performance computing (HPC) systems. Based on the strategic vision defined by University leadership and campus partners, the Senior Director of Advanced Systems sets the strategic roadmap for advanced HPC systems, provides technical leadership to their team and campus partners, and is ultimately responsible for cluster design, implementation, schedule efficiency, and administration. The Senior Director monitors research and technology trends in the computational research space and spearheads the technical design and implementation of services that meet campus needs This requires the Senior Director to remain current with trends in data science, machine learning, deep learning and AI technologies while anticipating faculty needs and strategically advocating for growth in Princeton-offered services. Responsible for understanding changes to campus need and technology evolution, the Senior Director of Advanced Systems plays a crucial role in balancing institutional investment with campus demand and new technology trends to meet the needs of our world class research community.

Requirements

15+ years of strong experience designing and managing high performance computing scheduling, hardware, software and storage systems.
Strong knowledge of current research computing hardware, software, and storage in the research computing industry, and the ability to stay abreast of trends and future HPC and AI systems and hardware.
Strong proficiency in SLURM scheduling and workload management software.
Extensive Linux system administration experience.
Networking skills (including Ethernet and Infiniband).
Demonstrated ability to work well with a diverse group of people at many different levels, including faculty, research staff, postdoctoral fellows, graduate students, undergraduate students, senior administration, IT staff (both at Princeton and at peer institutions).
Ability to facilitate productive conversation with staff through conflict, difference of opinion, and/or change in task prioritization.
Experience building cohesive and effective teams by promoting collaboration, self-reflection, and growth.
Ability to develop and maintain positive relationships with vendors.
The ability to manage and lead IT staff both through direct supervisory relationships and indirectly.
Strong communication skills, including written, verbal and presentation skills: effective in a variety of formal presentation settings: one-on-one, small and large groups.

Nice To Haves

Experience working in an Academic Environment
AI/Machine Learning/Deep Learning/Secure Research Infrastructure experience.

Responsibilities

Technical Leadership:
Provide vision, technical expertise, and direct support for computational research needs in academic departments across the university, including faculty, professional research staff, postdocs, graduate students, and undergraduate students.
Participate in and, in some cases, lead committees related to the governance and continued evolution of computational research clusters and services necessary to meet the needs of world class researchers at Princeton.
Monitor research and technology trends in high performance computing and research computing support by reading literature and white papers, following relevant periodicals, participating in related forums, and attending relevant conferences.
Remain current with the latest advancements in technologies supporting data science, machine learning, deep learning, and AI.
Develop and maintain strong and positive vendor relationships to keep abreast of technical advances and act as a primary point of contact for all interactions.
Work collaboratively with faculty, academic departments, research computing leadership, and senior administration to develop, evolve, and execute a coordinated strategic direction for deploying, operating, and supporting computational and data resources that meet the evolving needs of research and scholarship at the University.
Fully comprehends the challenges facing users of the computational cluster. Monitor ticket queues, coordinate support, and escalate troubleshooting of complicated problems.
Leverage team members' strengths to adapt to new challenges and requirements.
Strategic Vision and Deployment of HPC systems:
Spearheads the technical design and deployment of centralized research computing systems, consulting with faculty to understand their needs, interfacing with vendors to match capabilities to requirements, and leading the technical team to implement systems that provide the best balance of institutional investment and computational performance.
Lead the system administration for Princeton's centralized HPC systems that represent a significant capital investment to support production computational and data science research on campus and to enable researchers to scale to larger regional and national systems. The central HPC systems serve a large percent of the University's faculty and researchers from a majority of the academic departments.
Develop short and long term strategies promoting the growth and support of high performance computing and research computing systems.
Cluster Efficiency and Administration:
Utilize scheduling and workload management software (currently SLURM) for job submission, resource allocation, and monitoring within an HPC environment.
Design and optimize code for pre-processor and batch scripts for efficient execution of computational workloads.
Implement, maintain, and monitor fairshare policies to understand job queue flows and dynamically tune schedulers to improve efficiency.
Keep abreast of developments in scheduling and workload management software and its roadmap, and maintain strong relationships with software and workload management software developers and account managers.
Team Management:
Sets the direction for the Advanced Systems group; provides leadership and mentorship for staff, creates a positive and motivating work environment, sets clear goals, delegates tasks, monitors team performance, provides feedback and coaching, resolves conflicts, fosters strong interpersonal communication, and identifies opportunities for professional development.
Engage, collaborate, and align with colleagues in Researcher Engagement and Advanced Data and Storage Management to ensure shared awareness of campus needs and provide comprehensive support of services supported by all of Research Computing.
Promote a diverse, inclusive, and supportive work environment based upon a shared vision.