HPC Systems Engineer for Research Computing

The Hong Kong University of Science and Technology; Hong Kong; On-site; HKD 502,000 - 725,000; Full time

Benefits offered by this job

Annual leave

Medical benefits

Dental benefits

Gratuity on contract completion

Job summary

The Hong Kong University of Science and Technology (HKUST) information technology services invites applications for a position in the HPC group within the Enterprise Architecture and Technology team. The appointee will be responsible for system administration and day-to-day operation support of HPC clusters, including deployment, configuration, and maintenance of server nodes and cluster software.

Responsibilities cover job scheduling, software and MPI library support, container environments,

Qualifications

• Bachelor's degree in computer science, engineering, information systems or equivalent with at least five years of working experience.
• Sound knowledge of Linux system administration and related tools.
• Experience in an HPC or Linux cluster environment with job scheduling and resource management.
• Knowledge of NVIDIA GPU architecture and related software ecosystems is a plus.
• Good command of English and Chinese; strong analytical and teamwork abilities.
• Open to Associate IT Specialist for candidates with less experience (2-year contract).

Responsibilities

• Deploy, configure, and maintain server nodes, OS and cluster management software.
• Manage job queueing system and resource manager in a cluster environment.
• Install, update, and troubleshoot scientific applications, compilers, MPI libraries, and containerized environments.
• Manage high-speed ethernet drivers and parallel file system clients; troubleshoot IO bottlenecks; coordinate with network/storage teams.
• Handle user accounts, access permissions, security audits and firewall configurations.
• Provide consultancy support to students and researchers on HPC cluster usage.

Skills

Linux administration

HPC cluster management

Job scheduling systems

NVIDIA GPU knowledge

Networking for HPC

Security & access control

English & Chinese

Education

Bachelor's degree in CS/Engineering/IS or equivalent

Tools

Cluster management software

MPI libraries

Container technologies

High-speed networking tools

Job description

The Hong Kong University of Science and Technology (HKUST) information technology services invites applications for a position in the HPC group within the Enterprise Architecture and Technology team. The appointee will be responsible for system administration and day-to-day operation support of HPC clusters, including deployment, configuration, and maintenance of server nodes and cluster software.

Responsibilities cover job scheduling, software and MPI library support, container environments,