HPC Engineer

University of Maryland; College Park (MD); On-site; USD 120,000 - 170,000; Full time

Job summary

University of Maryland's Division of Information Technology seeks an HPC Engineer to design, deploy, maintain, and operate enterprise HPC resources across Linux platforms. The role involves installing, configuring, documenting, and troubleshooting hardware and software, with frequent travel between College Park, MD, and McLean, VA.

You will collaborate with the HPC team to stay current with advances in HPC architectures, parallel file systems, schedulers, Infiniband and Ethernet networking,

Qualifications

• Bachelor's degree in computing or a related field.
• Three (3) years of professional HPC engineering or systems design experience.
• Additional work experience may substitute for education on a year-for-year basis.

Responsibilities

• Design, deploy, maintain, and operate UMD's enterprise HPC resources.
• Install, configure, document, troubleshoot HPC systems and software.
• Collaborate with the national HPC community and present at conferences.
• Interface with industry HPC providers to stay current on architectures and tooling.
• Work with campus constituents to understand computing needs and ensure compliance.

Skills

Unix/Linux internals

HPC software stacks

Python/Perl/Shell

Infiniband networking

Ethernet networking

Parallel file systems

Automation/Config Mgmt

Cloud integration

Scripting languages

Education

Bachelor's degree

Tools

Docker

Singularity

Job description

Job Description Summary

Organizational Summary Statement:

This position is available within the University of Maryland's Division of Information Technology (DIT).

This position is for a High-Performance Computing (HPC) Engineer with the University of Maryland's Division of Information Technology (DIT). The Engineer participates in the design, deployment, maintenance, and operation of UMD's enterprise cutting-edge technology HPC resources. This includes installing, configuring, documenting, troubleshooting, and maintaining hardware, operating system software, and application software on a variety of platforms running Linux.

The hardware presently resides in College Park, MD and McLean, VA, and will require frequent travel between locations. The position entails collaborating nationally with the HPC community, staying in touch with emerging HPC technologies, and implementing best practices. Additionally, the Engineer, in partnership with the rest of the HPC team, will present at national conferences like SuperComputing and PEARC.

The Engineer also interfaces closely with industry solutions providers in the HPC ecosystem to keep abreast with developments in HPC architectures that work for the spectrum of research conducted at UMD from the social sciences to the natural sciences. The Engineer collaborates with campus constituents to understand their computing needs. Familiarity with HPC technologies such as parallel file systems, schedulers, Infiniband and Ethernet networking, cloud integration, automated configuration management, and large scale software deployments is highly desirable.

The Engineer should also be familiar with working with compliance frameworks, such as CMMC, NIST 800-171, and CUI. This position will occasionally require off-hours and weekend work for problem resolution and maintenance.

This position is considered essential and may be required to work at the normal work location or an alternative location during a major catastrophic event, weather emergency, or other operational emergency to help maintain the continuity of University services.

May be required to work evenings, nights, weekends, or different shifts for extended periods.

Physical Demands

Predominantly operates in an office environment. Some periods of time may be spent in computer rooms and communications equipment spaces for installation, testing, and troubleshooting. Some work (less than 5%) requires moving and lifting of heavy computing equipment. Must have the ability to walk short distances and drive a state vehicle.

Licenses/Certifications

Valid Driver's License

Minimum Qualifications

• Education: Bachelor's degree from an accredited college or university.
• Experience: Three (3) years of professional experience in high-performance computing (HPC) engineering or systems design.
• Other: Additional work experience as defined above may be substituted on a year for year basis for up to four (4) years of the required education.

KNOWLEDGE, SKILLS, & ABILITIES

• Knowledge of Unix/Linux operating system internals, client/server computing techniques, distributed systems management, and computer communications.
• Knowledge of current trends in commercial HPC solutions, including AI/ML architectures, accelerators, and federated authentication/authorization solutions (e.g., Shibboleth, LDAP).
• Skill in configuring and managing HPC software stacks, including network file systems (e.g., NFS, Samba) and open-source software compilation.
• Skill in scripting languages such as Python, Perl, and shell for automation and system management.
• Ability to install, modify, integrate and configure both commercial and open-source HPC applications.
• Ability to perform multiple tasks, prioritize workload and meet deadlines while demonstrating strong problem-solving and analytical reasoning.

Additional Job Details

Additional Information: Please note that all positions within the Division (DIT) have an in person component with expected time in our College Park, MD location per week.

Telework is not a guaranteed work arrangement.

Visa Sponsorship Information

• DIT will not sponsor the successful candidate for work authorization in the United States now or in the future.
• F1 STEM OPT support is not available for this position.

Preferences

• Demonstrated success in system administration and IT infrastructure design, implementation, and operations.
• Proficiency with HPC technologies such as parallel file systems, schedulers, Infiniband and Ethernet networking, cloud integration, automated configuration management, and large scale software deployments.
• Strong understanding of network and security protocols, policies, and standards.
• Proficiency with creating and using software containers (with, e.g., Docker, Singularity, or a similar tool), and/or creating workflows using these tools.
• Experience implementing/operating with controls for NIST SP 800-171, HIPAA Security and Privacy rules, or similar levels of security and