Improving Analysis of Large Multidimensional Data through Parallel Processing & Explorative Visualization

Using a combination of parallel processing and explorative visualization, a new and exciting solution for high-performance visual analysis of large multidimensional data is proposed. Steady volume flows are selected to demonstrate the strategy, methodology, techniques, and functionalities. Virtual Reality (VR) and Augmented Reality (AR) platforms will be adopted to unleash the power of explorative visualization. The proposed paradigm is well-suited for visual analysis of vast amounts of vector data, while it can be easily extended for scalar data visualization and information visualization. This project is part of a wider opportunity to establish more effective approaches for visualization of advanced data analytics. Our focus is on developing a solution with the following capabilities: • A scalable and flexible multi-threaded parallel processing system that can perform multiple terabytes worth of data processing tasks such as data parsing, pre-processing, organizing, data flagging, compressing, and reformatting • A data analysis environment using an off-the-shelf big data analytics engine that offers both explorative and predictive analytics capabilities to uncover hidden data relationships and patterns • A wide variety of visualization techniques to carry out algorithm-oriented scientific data visualizations and information visualizations. • A proposed cloud-based commercialization approach that meets all current DOD cybersecurity requirements. Approved for Public Release | 19-MDA-9932 (21 Feb 19)