HPC Engineer
What you will be doing
Playing a key role in defining and operating some of the most complex compute platforms that the client has to bring to bear against complex problems. These systems enable complex analysis, simulation and modeling leveraging massively parallel computing and disparate holding of very large data sets, to answer difficult questions. To do this you will assist the users in deploying jobs to these systems to harness the capabilities of these systems producing answers in the form of analytic product, models and simulations. This mission enablement is the heart of the hardest problems to solve.
• Responsible for the normal day-to-day HPC operations and maintenance of the HPC systems
• Provide day to day systems administration duties for Nvidia GPUs, Commodity Cluster Systems and Cray HPC environments
• Perform system monitoring, software installations, debug, upgrades, health checks, and identification/implementation of automated business processes
• Provide assessments, on-going performance analysis and recommendations for future architectures
• Responsible for operating all the host systems for the analysis
• Works in a liaison role, linking the analysts and their specialty codes and applications, to the computing systems that are focused on yielding in-depth technically sound results.
• Oversees analytic applications running on a clustered HPC fabric including CPU and GPU systems
• Managing job submission to clients applications and codes using MPI/OpenMPI
• Provide in-depth analytic results, to achieve a best-tool-for-the-job approach.
• Partners with data scientists, engineers, and analysts conducting specialized scientific and engineering analysis.
• Escalate issues and problems to hardware support and/or engineering management as necessary
• Responsible for continuous performance analysis and tuning the HPC environment
• Assist with the identification, troubleshooting, and repair of software problems impacting performance of implemented HPC solutions
• Perform installation of software patches including upgrades to operating systems and firmware
• Assist with the resolution of trouble tickets and software problems identified by system’s users
• Identify and expand services and functionalities offered in HPC environment
• Be a primary point of contact to resolve any hardware or software malfunctions, including working with service personnel as necessary
• Review system logs to identify and resolve software and systems related issues
• Prepare reports related to the operational efficiency of the hardware and execution of users jobs
• Experience with MPI/OpenMPI, SLURM, and Linux Operating Systems essential
• Prior experience as a Systems Administrator essential, with a preference for experience working with clustered systems including GPUs in the hardware stack
• Experience with high speed networking, and CUDA preferred
• Software integration experience a plus
• Other duties could be required to support the customer’s mission
What you will need
• Minimum of 6 years demonstrated on-the-job experience
• Demonstrated on-the-job experience with integrating functionality from disparate systems via scripting/tooling/automation
• Demonstrated on-the-job experience with the Sponsor's system security environment and requirements
• Demonstrated experience leading systems architecture, operations, maintenance and administration
Clearance:
Active TS/SCI with an appropriate polygraph is required to be considered for this role
