Job Description
Join the National Supercomputing Centre (NSCC) Singapore, a national research infrastructure managed by the Agency for Science, Technology and Research (A*STAR). We are seeking an experienced HPC Middleware Engineer to design, deploy, and optimise high-performance computing (HPC) middleware that powers cutting-edge scientific and engineering simulations. In this role, you will bridge the gap between hardware accelerators, parallel file systems, and end-user applications, ensuring seamless integration and peak performance for Singapore’s research community.
You will work alongside world-class researchers and system architects to maintain and enhance the software stack on the NSCC’s petascale supercomputers, including job schedulers, resource managers, containers, and performance monitoring tools. If you thrive on solving complex system-level challenges and have a passion for advancing scientific discovery through technology, this is your opportunity to make a tangible impact.
Responsibilities
- Design, implement, and maintain HPC middleware components such as job schedulers (Slurm, PBS), resource managers, and parallel file systems (Lustre, GPFS).
- Integrate container technologies (Singularity, Docker) and virtualisation solutions into the HPC environment for reproducible research workflows.
- Monitor and tune system performance, identify bottlenecks, and optimise data I/O, network, and compute resource utilisation.
- Develop automation scripts and configuration management tools (Ansible, Puppet) to streamline software deployment and upgrades.
- Provide technical support and consultation to NSCC users, assisting with job submission, debugging, and application performance profiling.
- Collaborate with system architects and hardware vendors to evaluate new technologies and plan capacity upgrades.
- Document system configurations, procedures, and best practices; contribute to user training and knowledge base articles.
- Participate in on-call rotation to ensure high availability and rapid incident response for critical HPC services.
Qualifications
- Bachelor’s degree in Computer Science, Computer Engineering, or a related field; Master’s degree preferred.
- At least 3–5 years of hands-on experience in HPC system administration or middleware engineering.
- Strong proficiency in Linux/Unix system administration, shell scripting (Bash), and at least one programming language (Python, C/C++).
- Deep understanding of HPC job schedulers (Slurm or PBS), resource managers, and parallel file systems (Lustre, GPFS).
- Experience with performance profiling tools (Intel VTune, perf, Darshan) and debugging (gdb, strace).
- Familiarity with containerisation (Singularity/Podman) and infrastructure-as-code tools (Ansible/Terraform).
- Excellent problem-solving and communication skills, with the ability to work collaboratively in a multidisciplinary team.
- Knowledge of high-speed interconnects (InfiniBand, OmniPath), MPI, and GPU programming (CUDA) is a strong plus.