Responsibilities:
System Administration & Maintenance:
Install, configure, and maintain HPC clusters (hardware, software, operating systems), perform regular updates/patching, manage user accounts and permissions, and troubleshoot/resolve hardware or software issues.
Performance & Optimization:
Monitor and analyse system and application performance, identify bottlenecks, implement tuning solutions, and profile workloads to improve efficiency.
Cluster & Resource Management:
Manage and optimize job scheduling, resource allocation, and cluster operations using tools such as Slurm, LSF, Bright Cluster Manager /
Base Command Manager
, OpenHPC, and Warewulf.
Networking & Interconnects:
Configure, manage, and tune Linux networking (TCP/IP, DNS, routing) and high-speed HPC interconnects (InfiniBand, Ethernet) to ensure low-latency, high-bandwidth communication.
Storage & Data Management:
Implement and maintain large-scale storage and parallel file systems (Lustre, Ceph, GPFS), ensure data integrity, manage backups, and support disaster recovery.
Security & Authentication:
Implement security controls, ensure compliance with policies, and manage authentication and directory services such as LDAP and Active Directory.
DevOps & Automation:
Use configuration management and DevOps practices (Ansible, Terraform, Jenkins, Git) to automate deployments, application packaging (RPM/DEB), and system configurations.
User Support & Collaboration:
Provide technical support, documentation, and training to researchers; collaborate with scientists, HPC architects, and engineers to align infrastructure with research needs.
Planning & Innovation:
Contribute to the design and planning of HPC infrastructure upgrades, evaluate and recommend hardware/software solutions, and explore cloud-based HPC solutions where applicable.
Qualifications:
- Bachelor’s degree in Computer Science, Engineering, or a related field (equivalent experience may substitute for degree)
- Minimum of 10 years of systems experience, including at least 5 years working specifically with HPC
- Strong knowledge of Linux operating systems (e.g., Rocky Linux, Ubuntu) with a fundamental understanding of Linux internals, system administration, and performance tuning
- Experience building and managing RPM and DEB packages
- Experience with cluster management tools such as Bright Cluster Manager, OpenHPC stack, or Warewulf
- Proficiency with job schedulers and resource managers such as Slurm and LSF
- Strong understanding of Linux networking (e.g., TCP/IP, DNS, routing) and HPC interconnects (e.g., InfiniBand, Ethernet) including performance tuning
- Knowledge of parallel file systems such as Lustre, Ceph, or GPFS
- Working knowledge of Linux authentication and directory services such as LDAP and Active Directory
- Proficiency in scripting languages (e.g., Python, Bash, R) and familiarity with MPI libraries for parallel and distributed computing (nice to have)
- Strong experience with DevOps and configuration management tools, including Ansible, Terraform, Jenkins, and Git
- Knowledge of HPC in cloud environments (e.g., AWS, Azure, GCP HPC offerings) is a plus
- Strong knowledge of Linux security, compliance standards, and data protection best practices
- Excellent communication, interpersonal, and problem-solving skills