Lead Systems Administrator
The Lead System Administrator is responsible for the ongoing operation, management, support, and security of Temple University computer systems running the Linux operating system. This is a technical position requiring a deep knowledge of the system administration skills required to insure our High-Performance Computing (HPC) infrastructure satisfies the requirements and service levels of the University. It will require ongoing consulting with departments, schools, and colleges, as well as external users and vendors, to guarantee successful project outcomes. The focus of this position is the support of competitive research and scientific computing. Some of the required skill set can be learned on the job, but a solid familiarity is required. The position will spend significant time insuring technical questions sent to our troubleshooting queue or email are addressed and resolved, while helping the researches to support themselves in our environment. The successful candidate will provide technical leadership and project management skills as well as assist and mentor other members of the team. The candidate will need to ensure the security of the computing environment by having a clear understanding of threats facing large computing facilities. Performs other duties as assigned. Required Education & Experience: Bachelor's degree in related field and at least three years progressive technical experience, including a solid background in Linux based systems. Demonstrated experience scripting for system administration and research purposes, including data manipulation and summation. Experience with EDR InfiniBand fabric networking turned for high performance and interaction with NFS client/server interactions. Experience using general GNU compilers, several different simultaneous versions of Java, Python 3.x, R, Sage math, Scala compiler and interpreter, and Apache Spark (in standalone mode). An equivalent combination of education and experience may be considered. Required Skills & Abilities: * Strong Analytical skills and experience in a fast-paced environment with different types of customers. * Ability to examine and improve the user interaction with HPC components and simplify the end-user experience on our platforms. * Attention to detail and ability to focus on critical projects as needed. * Knowledge of data transfer and data management, including transferring files and directories using ssh protocols such as scp, sftp, and rsync. * Robust understanding of Storage Area Networks design and troubleshooting, including shared file systems such as IBM GPFS. * Understanding of the batch job submission processes suitable for HPC environments. * Skills to assist researchers with common programming editors such as Vim and Emacs. * Ability to support interactive high-performance computing as well, requiring some skills in Gaussian, Magma, Maple, Mathematica, Matlab, SAS, and Strata applications. * Knowledge of and ability to use effectively Intel C, C++, and FORTRAN compilers and performance libraries for Linux. * Knowledge of coding utilities such as cmake, git, gdb, valgrind and other dynamic analysis tools. Preferred: * The University uses VMware virtualization technologies and EMC SANs storage to instantiate instances of Linux systems, including but not limited to Red Hat, SuSE, Ubuntu, Fedora, Debian, CentOS, Puppy, and others as needed for specific projects. A general familiarity with standing up and supporting such systems in an enterprise environment is desirable.