To be held in conjunction with
10-13 December 2020
The purpose of the Workshop is to bring together researchers, developers and practitioners in data science and high performance computing with domain experts to identify common interests, requirements, capabilities, and solutions in the analysis of large-scale, disparate data. With the proliferation of sensors in our environment and the battle space, data is being collected at unimaginable rates. Analysts are inundated with massive amounts of data, a small fraction of which may contain items of significance. Collecting data is less of a problem, while managing and gleaning value from that data has become a daily crisis. Our goal is a use-inspired, user-centric, and data flow oriented analytics process, and a common framework for training, evaluating, integrating, and transitioning analytics capabilities. Especially critical is the ability to respond to rapid, even daily, changes in threats, opportunities, data sets and algorithms with a commensurate ability to train, test, and deploy improved tools and to actually perform this at the tactical edge, i.e., at the point of need. The work should be guided by use patterns yet aimed to diminish the bottlenecks, minimize data movement and the number of discrete steps for the user, reduce the end-to-end time to complete a task, and reduce the cognitive workload on the user. Research is required to harness software frameworks, analysis tools, dynamic data sets, adaptive and high performance computing, and containers to produce a converged solution.