HPC Cluster Administrator III job vacancy

Vacancy expired!

Administrator Unique III (FIII3) - North Chicago, IL |

Job

# ABBVJP13377

AbbVie | w2 contract | 6 months (possible extension)

40/hr work weeks | Mon-Fri | Business Hours (on-site 3-4 days & remote 1-2 days per week)

Responsibilities & Duties

Primary responsibility: supports and administers on-premises Linux resources, including HPC Linux clusters and scientific Linux desktops. 2 on-premise HPC clusters that we're looking for someone to assist with. 1 with 50 nodes and another with 8 nodes that's going to be refreshed at the end of the year. Looking for HPC expertise. Investigating issues that people report, assisting with application or software install on a cluster, monitoring of the clusters.
Specific responsibilities include:
Troubleshoot and resolve hardware, Operating System and application issues in HPC and desktop environments
Work with systems staff to enhance our Configuration Management infrastructure
Evaluate performance impacts of planned operating system changes
Update and expand existing systems monitoring capabilities
Develop automation tools for systems administration
Maintain account management procedures to support growing number of researchers and scientists
Provide technical support to researchers using HPC resources, troubleshoot problems and develop appropriate computational strategies
Consult and collaborate with scientist co-workers to determine best system configurations for applications

Requirements

QUALIFICATIONS & SKILLS

Minimum of 5 years RedHat, CentOS and/or Ubuntu Linux system administration experience
Minimum of 3 years scripting experience with Bash, Perl or Python
Minimum 3 years HPC Linux cluster administration, including job queueing and resource management, parallel filesystems and network management
Prior experience with configuration management tools, such as Ansible, Chef, Puppet
Prior experience with physical Linux hardware
NFS experience.

Preference

Experience with HPC job queueing systems such as SLURM or PBS
Experience with HPC management software such as Bright Cluster Manager
Experience managing parallel and cluster file systems such as NFS, BeeGFS, GPFS, Weka, or Lustre
Network management experience especially Infiniband
Experience with GPU compute in HPC and desktop environments
Experience with container technologies such as Docker and Singularity
Environment module experience with tcl modules or lua modules
Linux physical hardware, whether that be interacting with servers or workstations.
On the smaller Cluster, we have Bright Cluster Manager that is also what the replacement Cluster will be using, so having this experience is a plus.
Parallel file system knowledge.
BeeGFS experience
GPFS experience
Lustre or Weka experience.
GPUs experience.
Docker or singularity experience.
Environment Modules experience.

Attributes For Success

Provide technical expertise to improve HPC cluster management, performance, and resiliency
Analytical thinking and problem solving skills for effective troubleshooting and problem resolution
Ability to work both independently and as part of the team; flexibility in dealing with assignments and in working on several projects simultaneously
Ability to effectively communicate with people of diverse backgrounds and computer knowledge