-
University degree in Computer Science or relevant studies in a related field
-
Experience managing Linux-based systems in a research or academic environment
-
Several years of experience designing, deploying or operating Linux-based HPC infrastructure
-
Strong Linux system administration skills, including kernel-level troubleshooting, system boot process, networking, storage and performance tuning
-
Hands-on experience with HPC hardware such as compute nodes, GPU servers, storage appliances and high-speed interconnects
-
Practical experience with InfiniBand fabrics, subnet managers and high-performance networking
-
Experience with BeeGFS (or comparable parallel file systems such as Lustre or IBM Spectrum Scale/GPFS)
-
Experience administering Slurm or comparable HPC workload schedulers
-
Good understanding of server hardware, firmware, BIOS/UEFI configuration, RAID, storage technologies and data center infrastructure
-
Experience installing, cabling and maintaining physical servers and storage systems in a data center environment
-
Knowledge of power, cooling and capacity planning for HPC installations
-
Experience with business continuity, backup and disaster recovery concepts
-
Strong analytical and troubleshooting skills, with the ability to work across hardware, operating system, storage and networking layers
-
Excellent analytical skills and attention to detail of data center operation
-
Strong communications skills and collaborative approach.
The DZNE operates a central IT located in Bonn, which provides the central network, server and storage infrastructure, and services for the entire research center distributed across 10 sites, supporting the work of scientists and administration. We operate a virtualization infrastructure, with Linux and Windows servers. We are seeking an infrastructure engineer with a strong understanding of how HPC systems are built and operated from the hardware upwards. This role focuses on Linux operating systems, compute nodes, storage, networking (InfiniBand), GPUs and cluster infrastructure. Experience using HPC systems for scientific computing alone is not sufficient; we are looking for candidates with hands-on experience deploying, integrating and troubleshooting HPC infrastructure.
