Research Systems Administrator-HPC III

Weill Cornell Medical College•New York, NY

1d•Onsite

About The Position

Resolves critical technical problems and takes lead in mentoring junior staff. Designs complex architectural solutions for a growing and adaptive computational environment. Regularly interfaces with lateral technical and research groups to assist in their endeavors.

Requirements

Bachelors of Science degree preferably in a technical field (e.g., computer science, physics, math, chemistry, or engineering)
Approximately 4 years of relevant work experience.
Experience working in a scientific computing environment, particularly in an academic setting is required.
Experience with Infiniband networking is required.
Experience with GPU computing, particularly using the nVidia CUDA toolkit is required.
Experience architecting scientific computing clusters and their associated scheduling systems, such as SGE, PBS, or SLURM is required.
Experience architecting database-backed web applications, particularly managing Shiny services, is required.
High proficiency in the Linux operating system, including scripting languages such as bash, Ruby and Perl.
Proficiency in building, installing, and configuring a variety of open-source Linux software packages, especially with complex dependencies.
High proficiency in at least two programming languages, such as C++ or Java, Perl, Python.
A detailed knowledge of computer hardware, specifically, Peta-scale storage platforms.
Proficiency in HPC filesystems, Lustre, GPFS.
Proficiency of networking concepts and use of tools and protocols such as SSH, DNS, DHCP, and LDAP.
Ability to manage complex vendor relationships.
Demonstrated leadership and mentoring skills to effectively manage and develop staff members.

Responsibilities

Designs the architecture of the department’s Peta-Scale storage infrastructure.
Architects data lifecycle management and storage tiers, including off-site replication to cloud providers.
Develops highly reliable batch scheduling infrastructure.
Designs the architecture of the department’s High Performance Computing (HPC) infrastructure.
Trains and mentors junior and operational staff in monitoring and troubleshooting cluster performance issues.
Designs and oversees the installation of scientific computing software stacks and research workflows.
Assists in the development of, coordinates, and leads, training events for research users in the efficient use of cluster and storage resources.
Provides and oversees complex technical documentation for the research community.
Directly interfaces with investigators to assess requirements and implement tailored solutions for their problem sets.
Responsible for vendor relations and negotiations of commercial solutions.
Performs other related duties as co-developed with management.