HPC Admin

Job Description:

Job Description

 Responsible for server administration, evaluation, planning, configuration, installing,

deployment, tuning and troubleshooting multiple large scale clustered / highly available HPC

environments.

 Strong technical skills in the construction and operations of High Performance Computing

Cluster servers and Linux

 Strong technical skills including knowledge of cluster management software and administrative

processes.

 Good knowledge of SLURM schedulers & Cluster manager tools like ROCKSXCAT

 Good understanding of Parallel file system.

 Knowledge in Installation & Configuration of HPC weather applications like MOMS.

 Ability to deeply understand application functionality and technical documentation required.

 Strong verbal and written communication and interpersonal skills.

 Ability to effectively work on and prioritize concurrent projects.

 Ability to perform basic Unix/Linux system administration and proficiency in PowerShell

 Ability to develop customized scripts/procedures for routine administration tasks.

 Experience in managing/administering Linux and Windows server environments for scientific

computing.

Skills and Qualifications

 B.E./B.Tech in Computer Science or related degree area

 Experience in HPC technologies such as parallel/distributed files systems (e.g. Lustre, GPFS), high

speed interconnect fabrics (e.g. Infiniband, Omni-Path), and HPC batch scheduling software

suites (e.g. PBSPro, SLURM)

 A minimum of 3 years of hands-on Linux experience (e.g. RHEL, CentOS) and production

infrastructure support (e.g. networking, storage, monitoring, compute, installation,

configuration, maintenance, upgrade, retirement)

 Experience in system administration and technical support (e.g. installation, configuration,

maintenance, upgrade, retirement, problem resolution)

 Excellent technical, analytical, and communication skills

 Proficiency in technical writing and documentation of solutions

 Works well in a team environment.

 Self-motivated