HPC Systems Administrator

5C Data Centers USA, Inc.•Springfield, OH

48d•$120 - $135•Remote

About The Position

The HPC Systems Administrator supports the day-to-day operation, maintenance, and troubleshooting of HPC and AI infrastructure across data center and cloud environments. The role focuses primarily on infrastructure and hardware, including Linux operating systems, GPU servers, high-speed networking, storage connectivity, firmware, drivers, and out-of-band management. Working with senior Systems Administrators, Network Engineering, Data Center Operations, Deployment Engineering, vendors, and other technical teams, the administrator troubleshoots infrastructure issues, restores systems to service, performs routine maintenance, and helps improve the reliability of production HPC environments. This role is suited to someone with a foundation in Linux systems administration or data center infrastructure who wants to develop expertise in HPC, GPU computing, high-performance networking, and large-scale AI infrastructure.

Requirements

Approximately 5 years of experience in Linux systems administration, data center operations, infrastructure support, HPC, cloud infrastructure, or a related technical role; equivalent hands-on experience, education, or technical training may be considered.
Working knowledge of Linux operating systems and command-line administration.
Basic understanding of enterprise server hardware and common server components.
Experience troubleshooting hardware or operating-system issues using logs and diagnostic tools.
Familiarity with networking fundamentals, including IP addressing, DNS, interfaces, routing, and basic connectivity troubleshooting.
Familiarity with server hardware concepts, including CPU, memory, storage, PCIe devices, NICs, power supplies, and firmware.
Familiarity with out-of-band management technologies such as IPMI, Redfish, iDRAC, iLO, or equivalent tools.
Basic scripting or automation experience using Bash, Python, PowerShell, Ansible, or similar technologies.
Ability to follow technical procedures and troubleshoot problems methodically.
Ability to recognize when an issue requires escalation and clearly communicate collected findings.
Strong written communication and documentation skills.
Ability to work collaboratively across technical teams.
Ability to participate in an on-call rotation.

Nice To Haves

Exposure to GPU infrastructure or an interest in developing expertise with NVIDIA accelerated computing platforms.

Responsibilities

Support and maintain Linux-based HPC and AI computing environments, including large-scale NVIDIA GPU clusters, across bare-metal, virtualized, containerized, and cloud-hosted platforms.
Troubleshoot hardware, operating system, networking, storage, driver, and firmware issues; document findings; maintain runbooks and support procedures; and escalate complex problems to senior engineers.
Assist with installation, configuration, validation, monitoring, and troubleshooting of NVIDIA GPU infrastructure, including drivers, firmware, system software, and GPU health management tools.
Perform post-maintenance system validation, collect and analyze diagnostic data for GPU and infrastructure faults, and coordinate hardware replacement and RMA activities for GPUs, system boards, NICs, power supplies, and other server components.
Administer and support enterprise Linux environments, including system services, filesystems, networking, storage, authentication, permissions, SSH, DNS, and NTP, while following security and operating system standards.
Troubleshoot operating system, CPU, memory, device-discovery, and kernel-related issues; collect diagnostic data for system crashes, hardware failures, and performance incidents.
Troubleshoot server hardware and support firmware updates for BIOS, BMC, GPUs, drives, and other infrastructure components.
Use out-of-band management tools for remote administration, system health monitoring, hardware inventory, and log collection; coordinate with Data Center technicians on hardware replacements, break-fix activities, cabling validation, and post-maintenance testing.
Assist with troubleshooting Ethernet, InfiniBand, and RoCE connectivity, including NVIDIA/Mellanox network adapters and link, interface, MTU, packet loss, VLAN, RDMA, and fabric-related issues.
Collect and analyze network diagnostic data, escalate complex switching and fabric issues to Network Engineering, and validate network performance after infrastructure changes, firmware upgrades, cable replacements, and deployments.
Participate in production infrastructure incident response, troubleshoot issues, provide status updates, and escalate incidents according to support procedures and service-level agreements.
Document troubleshooting activities, findings, corrective actions, and resolutions; follow change-management processes, maintenance procedures, rollback plans, and on-call support requirements.
Collaborate with senior Systems Administrators and engineers in troubleshooting, technical reviews, and post-incident analyses.
Contribute to runbooks, knowledge-base articles, and support documentation; share findings and lessons learned; and continue developing expertise in Linux, GPU infrastructure, high-performance networking, storage, and automation technologies.
Participate in an on-call rotation.