• Support and troubleshoot cluster compute and network hardware for large Linux-based HPC systems connected via Ethernet, Infiniband or other networks.
• Perform basic networking tasks to interconnect servers or components of clusters for communication.
• Monitor state of HPC systems and initiate appropriate actions to maintain a high availability level of resources.
• Fulfill technology support requests from Flatiron Institute’s staff.
• Manage issue resolution cycle including creation and tracking of tickets, communication with vendors, and generation of summary reports.
• Work with hardware vendors to find resolution of hardware issues.
• Provide guidance and assistance to the SCC team on escalated requests and contact vendors for support as necessary.
• Assist the SCC Director with department reporting and operations
• Keep up to date with the data center technologies.
• Utilize the Bright Cluster Manager and other configuration management systems to maintain configuration state of clusters.
• Track all changes to configuration with Git or similar version control system.
• Write and execute shell scripts in support of systems management, log analysis and other system administration duties for multiple systems. Use devops tools to properly document said changes and commit them to revision control systems.
• Travel required to our data center in Secaucus, NJ 1-2x/week.

• Knowledge of structured cabling components and concepts (patch panels, fiber, copper)
• Experience with data center facilities components (Rack Power Distribution Units, UPS, single and 3-phase power, CDUs )
• Basic understanding of Ethernet and IP networking.
• Ability to write technical documentation in a clear and concise manner.
• Understanding of system performance monitoring and actions that can be taken to improve or correct performance.
• Knowledge of the design, development and application of technology and systems to meet business needs.
• General knowledge of other areas of IT.
• Understanding of how system management actions affect users and dependent / related functions. Particular understanding of how these actions affect multi-tenant and multi-node environments that are characteristic of HPC systems.
• Familiar with Linux systems administration, and Linux shell scripts.
• Experience using issue-tracking systems to manage and document problem resolution.
• Familiarity with cluster management software such as Bright Cluster manager preferred
   

Related Skills & Other Requirements

• Detail-oriented and responsive to requests
• Patient, friendly and willing to find reasonable solutions
• Excellent interpersonal, verbal and written communication
• Self-motivated, works independently and as part of a team. Demonstrates problem-solving skills. Able to learn effectively and meet deadlines.
• Must be able to work in a data center, moving and working with rack-mounted machines.