DIBBS CIF21: Middleware and High-Performance Analytics Libraries for Scalable Data Science

Project Contact: Anil Vullikanti / Madhav Marathe
Funding Agency: National Science Foundation National Science Foundation

This collaborative project is supported by scientists from a variety of domains, including network science, epidemiology, spatial GIS, biomolecular simulations, pathology, computer vision, and remote sensing. This project addresses the need for high-performance data analytics with efficient parallel algorithms using a novel approach, centered on combining the breadth and productivity of best practice commodity Apache Big Data Stack (HPC-ABDS) and high-performance computing. The project will produce two types of key building blocks: Middleware for Data-Intensive Analytics and Science (MIDAS) and the Scalable Parallel Interoperable Data Analytics Library (SPIDAL). MIDAS and SPIDAL are motivated and tested by the applications in different domains.