Senior HPC Systems Engineer: Clusters & Storage

Fujitsu Limited; Singapore; On-site; SGD 120,000 - 180,000; Full time

Job summary

Fujitsu Limited is seeking a seasoned HPC administrator to manage the day-to-day operations of a high-performance computing environment in Singapore. You will ensure high availability, stability, and peak performance while administering HPCM for cluster provisioning, monitoring, and lifecycle management.

You will also optimize HPC storage, deploy and maintain the PBS Professional workload manager, and troubleshoot the Slingshot interconnect to maximize efficiency.

Qualifications

• 5–7 years of hands-on HPC administration experience in enterprise, research, or academic settings.
• Familiarity with configuration management and automation tools is advantageous.
• Understanding of GPU computing technologies (CUDA, ROCm) and accelerator-based HPC environments is a plus.

Responsibilities

• Manage day-to-day operations of the HPC environment ensuring high availability, stability, performance and reliability.
• Administer and maintain HPCM for cluster provisioning, monitoring, health management, software deployment and lifecycle management.
• Manage and optimize compute infrastructure delivering high performance and efficient resource utilization.
• Administer and optimize the Lustre parallel file system with large storage capacity and high I/O performance.
• Configure, administer and maintain the PBS Professional workload manager including queues and scheduling policies.
• Manage and troubleshoot the Slingshot interconnect fabric to ensure efficient communication between compute nodes and storage systems.
• Collaborate with infrastructure, storage, networking and application teams to support HPC deployments, upgrades and production operations.
• Provide technical support to researchers, scientists and engineers and assist users in optimizing HPC applications.

Skills

HPC administration

Ansible

xCAT

Bright Cluster Manager

Infrastructure-as-Code

CUDA/ROCm GPU computing

Tools

HPE Cluster Manager (HPCM)

PBS Professional

HPE Slingshot interconnect

ClusterStor Lustre

IBM Storage Scale (GPFS)

Job description

Fujitsu Limited is seeking a seasoned HPC administrator to manage the day-to-day operations of a high-performance computing environment in Singapore. You will ensure high availability, stability, and peak performance while administering HPCM for cluster provisioning, monitoring, and lifecycle management.

You will also optimize HPC storage, deploy and maintain the PBS Professional workload manager, and troubleshoot the Slingshot interconnect to maximize efficiency.