HPC System Administrator (Top Secret)

RedLine Performance Solutions, LLC.; Dayton (OH); Hybrid; USD 90,000 - 130,000; Full time

Benefits offered by this job

PTO

401k match

Health insurance

Job summary

RedLine Performance Solutions, a long-standing HPC solutions provider, seeks an experienced HPC Systems Administrator for on-site Dayton, OH. This role requires 24x7 on-call rotation and close collaboration with customers and the RedLine Program Manager to translate requests into actionable tasks.

Candidates must hold an active DoD Top Secret clearance and relevant technical certifications. The position offers full benefits, including PTO, 401k match, and comprehensive health coverage.

Qualifications

• 7+ years of Linux systems administration in Red Hat and/or Rocky environments.

Responsibilities

• Provide 24x7 on-call support for on-site HPC clusters.

Skills

Linux systems administration

TCP/IP networking

HPC cluster administration

Bash scripting

Perl scripting

Python scripting

Job scheduling software

Cluster automation tools

Lustre

InfiniBand

Communication skills

Organizational skills

Coordination across teams

Education

Active DoD Top Secret clearance

Linux+/Security+ certifications

Red Hat Certification (RHCSA/RHCE)

Tools

Slurm

PBS

xCAT

HPCM

Bright Cluster Manager

MPI

Warewulf

Weka parallel file systems

Job description

RedLine Performance Solutions (RedLine) has been in the High Performance Computing (HPC) solutions engineering services business for over 26 years and is consistently determined to keep the \"bar of excellence\" quite high for new hires. This enables RedLine to accomplish what other firms cannot and promotes a high level of staff retention. RedLine provides IT infrastructure management and technical support services to some of the world's largest supercomputing sites.

The HPC System Administrator will provide operational support for HPC clusters located in Dayton, OH. Operations run 24x7 and therefore there will be a rotational on-call requirement. The HPC systems administrator will engage with the customer and participate in the evolution and maintenance of the technical infrastructure in addition to operationally supporting the on-site HPC environment. The administrator will be responsible for relaying these insights to the RedLine Program Manager and working together to translate customer needs into actionable project tasks.

The position is at the customer site. The administrator functions as the lead point of contact for day-to-day operations and real-time problem resolution. As such, remote work is not viable for this role.

Active DoD Top Secret security clearance and relevant technical certifications (e.g., Linux+, Security+) are mandatory requirements for this position. This full-time (W-2) position offers a full benefits package including paid time off, 401k match, and health care benefits.

Required Skills:

• 7 or more years of Linux systems administration, preferably in a Red Hat and/or Rocky environment
• Strong knowledge of TCP/IP networking.
• 5 or more years of HPC cluster system administration experience, preferably with Dell clusters
• Strong experience in Bash, Perl, and Python scripting in a version-controlled environment using Git
• Experience with job scheduling software (e.g., Slurm, PBS)
• Experience with cluster automation tools (e.g., xCAT, HPCM, Bright Cluster Manager)
• Experience with parallel filesystems (e.g., Lustre)
• Experience with high-speed interconnects (e.g., InfiniBand)
• Strong verbal and written communication skills, with the ability to coordinate between multiple team members in remote locations between several disparate projects
• Strong organizational skills

Preferred Skills/Experience:

• Experienced with system engineering in addition to system administration
• Red Hat Certification (e.g., RHCSA, RHCE)
• Server automation experience (e.g., Puppet, Foreman, Ansible)
• Experience with MPI technologies.
• Experience with Warewulf cluster management and provisioning.
• Experience with Weka parallel file systems.
• Optimization experience with GPU based HPC clusters.