The Research Computing team is seeking applicants for a Junior Cluster Administrator to assist in providing day-to-day systems support to the Thorny Flat High Performance Computing (HPC) cluster at West Virginia University. The Junior Cluster Administrator will install, update, and support all scientific applications on the cluster, focusing on adapting codes to run on parallel architectures as well as taking advantage of our GPU hardware accelerators. As WVU has a wide range of STEM disciplines and codes utilized in our user base, the Junior Cluster Administrator must have a broad background in programming languages and experience in Linux-based operating systems.
In order to be successful in this position, the ideal candidate will:
Provide installation, updates, and support for all scientific applications on the cluster.
Provide routine and on-going systems maintenance and upgrades Thorny Flat High Performance Computing Environment.
Respond to system issues as they arise, which may occur after typical business hours.
Monitor and evaluate performance and operational integrity of cluster master nodes and computer nodes, and perform adjustments, configuration changes, and modifications necessary to assure the continued operation integrity of the systems and optimize their performance.
Provide routine, ad hoc, and as-needed reports to program managers on the operational integrity and performance of the resource and on user service-related issues associated with the operation, use, and support of the cluster.
Interact and collaborate with other administrators affiliated with Informational Technology Services and technical staff at the Pittsburgh Supercomputing Center and the XSEDE program.
Bachelor's degree in computer science or related field with specific coursework in programming languages and operating systems;
Minimum of two (2) years of experience in Linux-based cluster and scientific high-performance computing systems operations and management;
Or an equivalent combination of education and experience.
Must meet ITAR definition of U.S. Person.
HPC cluster administration and setup skills.
Linux system administration skills.
Working knowledge of software packages related to cluster computing, but not limited to, PBS/TORQUE, Moab Cluster, Suite, SLURM, MPI/MPICH/Intel Parallel Studio, common Linux services (Apache, NFS, LDAP, etc.).
Basic network knowledge of a network environment including managing Layer 2 Ethernet Switches.
Knowledge on scientific programming languages such as C, FORTRAN, CUDA, Python, R, MATLAB, etc.
Knowledge of RAID concepts and management.
Knowledge of Parallel File systems including IBM's General Purpose File System (GPFS) and/or Lustre.
Working knowledge of revision control systems such as 'git'.
Ability to work effectively as team member.
Ability to read and interpret documents such as safety rules, operating and maintenance instructions, and procedure manuals.
Effective communication skills.
West Virginia University Research Corporation Human Resources