HPC Administrator

Responsibilities:

Run:

• Workload scheduler management
• HPC tools and middleware management
• HPC Application troubleshooting from infra perspective
• System/cluster monitoring
• Management, monitoring and maintenance of InfiniBand interconnect, cluster services, cluster hardware
• Root Cause Analysis of HPC cluster

Build:

• System deployment using cluster management tools
• OS repository management along with compatibility matrix for various device drivers
• Configuration and maintenance of network services
• Deployment and management of monitoring systems to automate services and hardware alerts
• Internal cluster network connectivity
• Cloud migration of HPC cluster (Core HPC system)
• Installation and configuration of HPC workload managers
• HPC Application integration with job scheduler

Skills / Expertise:

1.Operating systems: Linux: RHEL, Rocky, CentOS, SuSE, Windows

Schedulers & Resource Managers: PBS Pro, LSF, SLURM, Open Grid Scheduler [OGS]

2.Provisioning: HP-CMU, xCAT, Bright Cluster Manager

3.Monitoring: Ganglia, Nagios, Zabbix, Grafana

4.Configuration Management: Chef, Puppet, Ansible, CFEngine.

5.HPC Application: Openfoam, StarCCM+,Abaqus, Ansys, Ls-Dyna and other CAE & CFD applications

6.Linux operating system fundamentals, architecture, administration, native service configuration and advanced debugging skills

7.Knowledge of x86 hardware, system software and system services

8.Experience in HPC cluster configuration, management, upgrade and migration

9.Knowledge of Managing parallel file system Like Luster, BeeGFS, GPFS

10.Scripting and automation – bash, Perl, Python

11.Knowledge of ITSM processes