HPC System Administrator

HPC System Administrator

Requirements

• Must be a US Citizen or hold a Permanent Resident Card.
• Ability to obtain a Public Trust
• Bachelor’s Degree or FOUR (4) years of additional equivalent experience in lieu of a degree.
• A minimum of SIX (6) years of experience in Linux HPC systems administration, and less experienced candidates with outstanding qualifications will also be considered.
• Extremely competent with using and managing an exclusively Linux environment, including desktop support.
• Experience managing resources in an HPC environment.

Responsibilities

• Oversee various components of the LoBoS cluster remain in good working order such as network configuration, firewall management (Palo Alto), file system management (ZFS, VAST), security, batch queuing systems (SLURM), database administration, distributed computing, file transfer services, web servers, and electronic mailing lists.
• Serve as a technical resource for HPC, LCB, NHLBI, and other NIH personnel in areas such as the Linux operating system, networking, database system administration, distributed computing.
• Oversee configuration and installation of virtual and physical servers and manage upgrades to existing hardware.
• Ensure patches, security updates, and configuration changes to software systems are applied to enhance reliability and to meet security needs.
• Assist in maintaining the LoBoS Assessment & Authorization package based on National Institute of Standards and Technology (NIST) SP 800-53 security controls under guidance from NHLBI’s Information System Security Officers (ISSO).
• Stay informed regarding new developments in hardware/software and evaluate their potential usability for LoBoS/LCB.
• Evaluate the existing system to determine when updates/upgrades to hardware and/or software are necessary.
• Manage the budget used to procure new hardware/software for LoBoS.
• Prepare software documentation and technical reports related to assigned projects.
• Collaborate with Office of the Chief Information Officer (OCIO), Center for Information Technology (CIT), and NHLBI security teams to ensure adherence to compliance policies.
• Participate in conferences and meetings of professional groups concerned with the application of HPC, AI/machine learning, and other emerging computer technologies.

Preferred Qualifications

• Experience implementing and managing SLURM batch queueing software preferred.
• Extensive knowledge of at least two high level computer languages such as C, C++, FORTRAN, Ruby, Perl, or Python is desirable.
• Comprehensive knowledge of shell scripting.
• Broad knowledge of systems administration tools such as Puppet, Ansible, etc.
• Familiarity with logging tools (such as greylog, nagios, etc.) and management tools (Ansible, puppet/chef).
• Experience with government computer security rules and standards is desirable.