The incumbent will receive assignments in the form of objectives with goals and the process by which to meet goals. S/he will provide direction to staff according to established policies and management guidance, administer policies that directly affect subordinate staff, recommend changes to department policies and practices, and identify risks and responds accordingly. Additionally, the incumbent will provide priority setting and work flow analysis.
The High-Performance Computing (HPC) Systems manager provides supervisory, technical leadership and business acumen in running the High-Performance Computing Systems Group. The manager serves as the group's liaison to SDSC, Data-Enabled Scientific Computing (DESC) division, project management and through them to the user community that DESC serves. The manager's primary responsibility is to maintain the production HPC and data systems, and supports research by contributing in the areas of cyberinfrastructure systems architecture and design, and technology assessment. The manager tracks emerging technologies, and contributes to the technology evaluation and design for new systems, including contributing to proposals for such systems. The manager is expected to promote SDSC’s capabilities, and give talks at technical venues in areas that are central to the HPC system group.
The HPC Systems manager is responsible for the deployment, operation, and maintenance of SDSC’s high-performance Linux-based systems, including high-performance clusters (currently Comet, Gordon and TSCC) and their parallel file systems (e.g. the multi-petabyte Data Oasis), as well as the overall operating system environment, networking, and associated software. Beyond supervising the HPC Systems Groups this includes handling maintenance engagements and negotiating support contracts with vendors and suppliers. Working with SDSC's User Services, Scientific Computing Applications group and XSEDE Extended Collaboration Support Services (ECSS) team the manager is responsible for supporting machines used by users from 100s of institutions. The manager is the primary liaison to the XSEDE Operations team, responsible for ensuring that SDSC's systems meet required standards of integration as a Service Provider to the XSEDE program. The manager will work closely with SDSC management, principal investigators, and project managers for the systems to ensure that SDSC and program objectives are met, and will coordinate extensively with other cyberinfrastructure groups in the center, including User Support, Storage Systems, Networking, Security and Facilities.
- Position requires on-call work.
For more information, please visit www.sdsc.edu.
Has solid knowledge of related principles and concepts to provide solutions to more complex and diverse information systems issues.
Proven knowledge of the applications programming development function.
Solid ability to gather, organize, and analyze data in the completion of assignments.
Advanced knowledge of Linux, including services, security, networking, and file systems.
Proven experience as an administrator or developer of large-scale HPC clusters.
Demonstrated experience of HPC interconnect and HPC parallel file system software.
Proven experience working independently in environment with frequent interruptions. Ability to prioritize and complete projects as scheduled.
- Job offer is contingent on a clear background check; will work on sensitive projects requiring security/background check.
The University of California is an Equal Opportunity/Affirmative Action Employer. All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, national origin, disability, age, protected veteran status, gender identity or sexual orientation. For the complete University of California nondiscrimination and affirmative action policy see: http://www-hr.ucsd.edu/saa/nondiscr.html