Job SummaryWith consultative direction within the Core Services group, this position is responsible for architecting, optimizing, and maintaining a high-velocity, on-premises Linux data pipeline supporting real-time, AI-driven machine learning models for the DIII-D national fusion reactor program.
As a hands-on technical practitioner, you will manage bare-metal deployments, orchestrate high-availability parallel storage arrays, and troubleshoot ultra-low-latency network protocols across a heavily segmented, secure infrastructure. This role focuses on deep, OS-level infrastructure and physical data center operations, bridging the gap between raw hardware capabilities and mission-critical scientific computing.
DUTIES & RESPONSIBILITIES - Bare-Metal & OS Administration: Plan, manage, and optimize day-to-day operations of on-premises, physical x86 and GPU server infrastructure running Red Hat Enterprise Linux (RHEL).
- High-Velocity Networking: Provision and troubleshoot core network sharing protocols (NFS) and ultra-low-latency networking hardware supporting 100GbE backbones and RDMA over Converged Ethernet (RoCEv2).
- Operational Technology (OT) Security: Maintain and fortify a heavily firewalled, push-only segmented internal network architecture, ensuring strict "Deny All Inbound" rules protect sensitive reactor control systems.
- Storage Architecture Lifecycle: Build, configure, and maintain high-density NVMe storage tiers and software-defined, parallel filesystems (BeeGFS/ZFS) to handle massive multi-gigabyte payload dumps with zero ingestion bottlenecks.
- Performance & I/O Benchmarking: Conduct granular system-level I/O benchmarking, diagnose deep kernel-level platform anomalies, and implement kernel parameter tuning to maximize data velocity and prevent data loss.
- Automation & Provisioning: Develop and maintain automated infrastructure workflows and configuration management frameworks using Bash shell scripting and Ansible to streamline bare-metal node deployment and environment validation.
- Data Center Operations: Manage physical data center footprints, including high-density server rack layouts, equipment delivery, hardware diagnostics, power infrastructure/UPS, and asset lifecycle tracking.
- Collaborative Consultation: Act as a technical infrastructure expert, guiding the development of innovative solutions to unique computing challenges for data acquisition systems and scientific teams.
- Vendor & Planning Coordination: Analyze new hardware architectures, engineer custom hardware Bill of Materials (BOM) alongside OEMs/ODMs, and represent the organization as a primary technical contact with suppliers.
- Compliance & Safety: Observe all laws, regulations, and facility safety obligations, ensuring that all system upgrades and physical data center modifications are designed with established personnel operating procedures properly considered.
Job QualificationsRequired Core Skills:
- Typically requires a bachelor's degree in information technology or a related discipline and fifteen or more years of progressive professional experience in an information technology department primarily in systems administration. May substitute equivalent working experience in the field in lieu
- Expert Linux Skills: Detailed and extensive technical expertise in Red Hat Enterprise Linux (RHEL) system design, installation, configuration, and low-level kernel troubleshooting.
- Storage Mastery: Proven hands-on experience implementing, configuring, and performance-tuning ZFS, RAID, and LVM storage structures.
- Networking Fundamentals: Comprehensive understanding of TCP/IP networking, host-based firewalls, network bonding, and debugging network sharing protocols across segmented topologies.
- Automation Background: Demonstrated proficiency writing advanced automation scripts in Bash or Python, coupled with configuration management toolsets (Ansible).
- Physical Infrastructure Competency: Direct experience handling physical server deployment, rack configuration, hardware component replacement, and diagnostic testing within an enterprise data center environment.
Preferred (Nice-to-Have) Pipeline Skills:
- Familiarity with High-Performance Computing (HPC) cluster environments and workload scheduling managers (e.g., SLURM, PBS Pro).
- Experience configuring or administering parallel file systems (e.g., BeeGFS).
- Familiarity with distributed, event-driven streaming architectures or message brokers (e.g., Apache Kafka).
- Familiarity with in-memory data structures or caching layers (e.g., Redis).
- Experience with open-source hypervisors and virtualization platforms (e.g., Proxmox).
- Exposure to GPU computing clusters and high-performance environments (e.g., NVIDIA/Mellanox).
