Track 2: Customized Multi-tier Assistance, Training, and Computational Help (MATCH) for End User ACCESS to CI

Abstract NSF-supported computational cyberinfrastructure (CI) has been highly successful in advancing science and engineering over the last few decades. Along with that success, there have been significant changes in the size and composition of the user community, the architecture and capacity of compute, storage, and networking platforms, and the methods by which researchers and CI professionals communicate among themselves and with each other. These changes require a rethinking of the role of user support services, and how they are delivered. To address these major changes, we propose a new model for implementing end user support services that leverages modern information delivery systems and simplifies user interfaces to provide cost-effective scaled support to a broader community running jobs on a wide range of cyberinfrastructure. The project will develop a hierarchical user support system designed (at the lowest level of the hierarchy) to address the vast majority of user support issues with automated and online information and communication services that leverage the broader community’s knowledge. We will also deploy common and simple to use interfaces that make it easy for users to run a wide range of frequently used applications, hiding and handling details on the user’s behalf. We also plan to leverage experts from the community to develop training materials and instructions that can dramatically reduce the user learning curve for several increasingly important CI computational techniques including artificial intelligence, machine and deep learning, big data analysis and processing, geospatial and time series data analysis, multimedia data analysis, as well as the associated data movement and storage challenges that arise when working with today’s big data. For projects requiring personalized or custom assistance, as well as specialized future systems (Quantum accelerators, Neuromoric processors, etc.), we will develop and employ a matchmaking service that will maintain a database of specialist mentors and student mentees that can be matched with projects to provide the domain-specific expertise needed