King Abdullah University of Science and Technology (KAUST)

Location: Thuwal
Salary: Competitive Tax-free Salary
Hours: Full Time
Contract Type: Permanent
Placed on: 12th December 2014
Expires: 12th February 2015

Job Location: Saudi Arabia
Job Type: Non-Academic

The KAUST Supercomputing Core Lab (KSL) is seeking a high performance computing System Administrator to administer KSL supercomputers, storage systems and other systems and services; Large-scale Linux system administration expertise, expert level configuration management, scripting, backup, monitoring and automation skills are required;

This is a highly technical role, not a management position;

The successful candidate will have experience in similar roles in high performance computing (HPC) labs or university settings.

Major Responsibilities

Be part of the team administering supercomputers, clusters, storage, and other resources described on the KSL Facilities page at the School Website
Participate in planning and deploying new resources, as well as retirement of older resources;
Apply expert knowledge of Unix/Linux systems administration, including all aspects of management, monitoring, performance analysis, and integration in complex heterogeneous environments;
Use configuration management tools (e.g., xCAT, puppet, IPMI) to help maintain large-scale Linux clusters, supercomputers, storage systems, and smaller systems.
Understand, implement, troubleshoot, and support batch and workload management systems, including diagnosis of failed jobs, implementation of KSL policy, and investigations of new features and services;
Plan for and deploy patches and updates to the operating system and application software;
Troubleshoot and solve difficult system issues as they arise, such as networking (InfiniBand, Ethernet LAN), filesystem issues (GPFS, NFS, Lustre, etc.), hardware, operating system, and application software;
Understand and support hierarchical filesystem infrastructure, software and services, including high performance parallel storage, backup systems, and robotic tape libraries;
Monitor, optimize services and performance (file system, network interconnects) using Nagios, Ganglia etc.;
Develop reports and customize tools that automate the monitoring process of critical systems and alert team of issues automatically;
Administer management servers for infrastructure (file servers, monitoring, etc);
Solve escalated systems related issues, coordinate with vendors to isolate hardware problems, install firmware or software patches as necessary;
Troubleshoot and tune high speed interconnects such as 10GbE, InfiniBand and fibre channel;
Act as point of contact for system related matters for departmental researchers and faculty

This is a highly technical role, not a management position;

The successful candidate will have experience in similar roles in high performance computing (HPC) labs or university settings.

Major Responsibilities http:///8HPdss