HPC Workload/Job Scheduler Administrator

Leidos•Bay Saint Louis, MS

13h•Onsite

About The Position

Leidos is looking for an Intermediate to Advanced level expertise to work in their HPC environment at the Navy DSRC team supporting the Department of Defense (DoD) High Performance Computing Modernization Program (DoD HPCMP). This is an on-site position. The successful candidate will support the day to day operations, testing, deployment, administration/management, reporting, and analysis tools for examination of workload management/job scheduler activity on high performance computers. Additional responsibilities are outlined below. Flexibility in approach to finding solutions is essential, as well as the ability to understand and act on the customer needs and priorities and learn new technologies quickly. Occasional off-hours activity will be required to perform updates and take part in service restoration efforts. Locations: Stennis Space Center, MS

Requirements

Candidates shall have a bachelor’s degree in computer science or related field and at least 8+ years of experience in a large and complex IT environment providing industry and government recognized functional expertise, or a master’s degree with 6+ years of experience, additional years of experience will be accepted in lieu of a degree .
Must possess a Top Secret clearance
IAT Level II Certification Required.
Accepted professional IAT Level II certifications include RHCSA or higher Red Hat certification, CompTIA Linux+ and Security+.
Red Hat Enterprise Linux (RHEL), CentOS, or Linux variants operating systems (OS)
Experience providing industry and government recognized functional expertise with the following skills:
Complies with software requirements and requests from the customer regarding workload management, including validation, scheduling policies, and post-run processing (mainly accounting).
Performs installation, testing and support for COTS, GOTS, and open-source software on high performance computing platforms as well as local infrastructure systems (when applicable) for the Navy DSRC.
Provides hands-on support and administration of Workload Management Batch Job Schedulers such as Altair PBS Pro, Slurm, with a potential to expand to other products.
Correctly forecast and express resource limitations and provide recommendations for increasing the efficiency of our resources through proper scheduling and load balancing techniques.
Provide after-hours support and respond to emergency situations and resolve problems.
Develops and monitors policies and standards for allocation related to the use of computing resources.
Responsible for the successful design, support, and integration of HPC clusters (computation, storage and infrastructure), software, scheduling, and research applications in order to meet the computational needs of DoD scientists.
Participates in the installation, integration, acceptance testing, and on-going maintenance of our HPC systems and software environment.
Installing, upgrading, and supporting Batch Job Schedulers.
Maintain and/or develop software code that is used to report Job Accounting on HPC systems to the HPCMP.
Fielding and processing issues from peers, users, or management. (would include articulating issues to third-party services and working through a problem through text.)
Enhancement and automation of managed software - Develop, install, and maintain requested software. This may extend beyond workload management software, including file/data profiling (archive scraping), text transposing/linters, and interactive processing scripts.

Nice To Haves

Portable Batch System (PBS), Slurm, or IBM Spectrum LSF schedulers.

Responsibilities

Provide Tier III HPC support to HPC site.
Complies with software requirements and requests from the customer regarding workload management, including validation, scheduling policies, and post-run processing (mainly accounting).
Performs installation, testing and support for COTS, GOTS, and open-source software on high performance computing platforms as well as local infrastructure systems (when applicable) for the Navy DSRC.
Provides hands-on support and administration of Workload Management Batch Job Schedulers such as Altair PBS Pro, Slurm, with a potential to expand to other products.
Correctly forecast and express resource limitations and provide recommendations for increasing the efficiency of our resources through proper scheduling and load balancing techniques.
Provide after-hours support and respond to emergency situations and resolve problems.
Develops and monitors policies and standards for allocation related to the use of computing resources.
Responsible for the successful design, support, and integration of HPC clusters (computation, storage and infrastructure), software, scheduling, and research applications in order to meet the computational needs of DoD scientists.
Participates in the installation, integration, acceptance testing, and on-going maintenance of our HPC systems and software environment.
Installing, upgrading, and supporting Batch Job Schedulers.
Maintain and/or develop software code that is used to report Job Accounting on HPC systems to the HPCMP.
Fielding and processing issues from peers, users, or management. (would include articulating issues to third-party services and working through a problem through text.)
Enhancement and automation of managed software - Develop, install, and maintain requested software. This may extend beyond workload management software, including file/data profiling (archive scraping), text transposing/linters, and interactive processing scripts.