CyberTraining: Pilot: HPC ED: Building a Federated Repository and Increasing Access through Cybertraining

The pilot will allow for the discovery, identification, and sharing of High Performance Computing (HPC) training materials across a multitude of providers and users. While there are considerable resources for learning about HPC topics online, the overall environment is highly fragmented and challenging to navigate. The project aims to create a federated repository of training materials with metadata, quality assurance, and roadmaps that can be accessed by both computational resource providers and computational scientists to discover and use these materials.

Training providers can list their resources through the repository (HPC ED) and make them accessible to a multitude of portals across the United States and beyond. Examples of organizations that will benefit are Kean University, a Hispanic-Serving Institution, which will train students to use their local computational cluster, and DesignSafe, a Natural Hazards Engineering Research Infrastructure facility, which will train their engineering community on how to use a multitude of HPC resources.

HPC ED will pilot a federated training repository that can be leveraged by members of the cyberinfrastructure community to provide access to the best training resources while maintaining relationships with their user base. Throughout the cyberinfrastructure community, many resources are available to train faculty and young scholars about the successful utilization of computational resources for research.

The problem that HPC ED addresses is that training materials abound, but with little information about the quality or relevance of offerings. The federated repository will identify and vet training resources from this broad set of offerings and provide metadata and characterization that support the successful discovery and utilization of training resources, and their incorporation into portals for research computing groups at universities, HPC centers, schools, domain-centered institutions, and elsewhere.

Local centers will be able to include content identified in the repository and offer it to their local faculty members with local branding and identification side-by-side with their locally developed offerings. In turn, their training offerings can be vetted and shared with the broader computational science community, increasing their reach and impact.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.