HPC High-Performance Computing System Administrator

D L Resources Pte Ltd; Singapore; Sur place; SGD 50 000 - 80 000; Plein temps

Avantages offerts par ce poste

null

Résumé du poste

D L Resources Pte Ltd in Singapore is seeking a hands-on HPC Administrator to support deployment, monitoring, and maintenance of High Performance Computing environments, plus core Linux-based infrastructure.

You will work with a small team to ensure compute and storage availability, assist with job schedulers, and help optimize performance through routine health checks and runbooks.

Qualifications

• 1–3 years in IT infrastructure or system administration.
• Basic exposure to HPC environments or cluster computing.
• Strong Linux knowledge (RHEL/Ubuntu/CentOS) and CPU-based servers.
• Understanding of networking and storage concepts.
• Scripting in Shell or Python preferred.

Responsabilités

• Support day-to-day operations of HPC clusters (compute, storage, networking).
• Monitor system performance, scheduling, and resource usage.
• Assist in Linux/Unix installation, configuration, and patching.
• Manage CPU-based servers and virtual environments.
• Support job schedulers (Slurm, PBS, LSF) at a basic level.
• Perform health checks, troubleshooting, incident resolution.
• Assist user onboarding, access management, environment setup.
• Collaborate with senior engineers on optimization and scaling.
• Maintain documentation and runbooks.

Connaissances

IT infrastructure

System administration

Linux knowledge

Networking fundamentals

Scripting (Shell/Python)

Troubleshooting

Outils

Slurm

PBS

LSF

Description du poste

Client Provided Role Summary:

The HPC Administrator will support the deployment, monitoring, and maintenance of High Performance Computing (HPC) environments, along with broader infrastructure components including CPU-based servers and operating systems.

Client Provided Key Responsibilities

• Support day‑to‑day operations of HPC clusters (compute, storage, and networking)
• Monitor system performance, job scheduling, and resource utilization
• Assist in installation, configuration, and patching of Linux/Unix-based systems
• Manage CPU-based servers and virtual environments
• Support job schedulers (e.g., Slurm, PBS, LSF – basic exposure)
• Perform system health checks, troubleshooting, and incident resolution
• Assist in user onboarding, access management, and environment setup
• Work with senior engineers on cluster optimization and scaling activities
• Maintain documentation and operational runbooks

Client Provided Required Skills & Experience

• 1–3 years of experience in IT infrastructure / system administration
• Basic exposure to HPC environments or cluster computing
• Strong knowledge of: Linux (RHEL, Ubuntu, CentOS) and CPU-based servers (bare metal / virtualized)
• Understanding of: Networking fundamentals (TCP/IP, DNS, etc.) and Storage concepts (NAS, SAN, parallel file systems – basic awareness)
• Scripting knowledge (Shell / Python – preferred)
• Good troubleshooting and analytical skills

Client Provided Preferred Skills

• Exposure to GPU-based environments (nice to have)
• Awareness of cloud platforms (AWS / Azure HPC offerings)
• Familiarity with monitoring tools

=============================

What an ideal candidate should have:

(A) Required / Basic Exposure:

• Understanding of HPC cluster architecture, including compute nodes, storage systems and networking components.
• Basic exposure to HPC job schedulers (Slurm, PBS or LSF).
• Familiarity with Linux-based HPC environments and system administration.
• Understanding of HPC workload monitoring, resource utilisation and troubleshooting.
• Basic knowledge of enterprise storage, filesystems and networking.

(B) Preferred / Good to Have:

• Exposure to parallel filesystems (GPFS, Lustre or BeeGFS).
• Familiarity with InfiniBand, RDMA or high‑speed cluster networking.
• Exposure to GPU‑based HPC environments (NVIDIA GPU/CUDA).
• Experience supporting HPC workload optimisation or performance monitoring.
• Exposure to HPC cluster deployment, provisioning or configuration.