Workshop: Directed Internship and Apprenticeship toward a Sustainable HPC Workforce Development Pipeline
Abstract: In high performance computing (HPC), staff who manage the computational center can consistently turn over and recruiting new staff can be a challenge. The skill sets required to perform this function involve understanding of system administration, storage administration, wide area network engineering and management of the facility environment. As such, this area intersects with concepts in computing science, electrical engineering and mechanical engineering, which is currently not taught in any educational system in this combination. This talk provides information on a training model that has been adopted at the National Energy Research Scientific Computing Center (NERSC) at Lawrence Berkeley National Laboratory (LBNL). As a computational facility that have onsite staff 24/7, staff are trained with required technical understanding and independent thought to resolve problems to bring the facility resources back to normal operation.
Most education systems focus on one or two of these areas. The model involves working with the school and curriculum to provide the basic education required. Students are placed in an internship to demonstrate their knowledge of these areas but also direct their training toward the HPC area. Completion of an internship can lead to the one-year apprenticeship program which is milestone based and meant to direct the apprentice to master each of the skill areas to prepare to meet the minimum qualifications of a Site Reliability Engineer at a computational facility. NERSC has seen success from this program and has created a workforce pipeline from schools into a potential career position.