AAG 2024 Annual Meeting
Authors: Devika Jain, Jeff Blossom and Wendy Guan, Harvard University
Abstract: In the rapidly evolving field of geospatial big data science, the demand for robust infrastructures capable of handling big data while ensuring cost efficiency and performance has become paramount. This infrastructure will encapsulate all essential components necessary for geospatial big data computation, encompassing data acquisition, storage, processing, analysis, visualization, and security. Our infrastructure is built using New England Research Cloud (NERC) and Harvard’s Research Computing Cluster (FASRC). It aims to provide GPU based real-time visualization solutions such as Heavy.ai for interactive visualization of billions of objects. It builds CPU-based big data computation solutions using ESRI’s GeoAnalytics Engine and PostGIS. It provides capabilities to handle a wide variety of datasets such as Social Media, Climate and WebAI. It includes containerized, scalable and reusable GIS solutions for complex analysis such as RINX for climate data analysis, and open-source geospatial datasets which can be used by larger academic communities such as Twitter Sentiment Geography Index. The emphasis is placed on building a high-performance infrastructure capable of handling both vector and raster big data. It utilizes scalable storage systems, distributed computing frameworks and cloud computing to harness the potential of geospatial analytics. Through a comprehensive analysis of infrastructure design, app development, case studies, workshops and demos, this project offers insights into best practices and strategies to develop resilient infrastructures tailored to the demands of geospatial big data analytics. It provides a guiding framework for researchers on designing, implementing, and managing cloud/cluster based infrastructures for geospatial big data.
LOCATION:Honolulu, HI STATUS:CONFIRMED DTSTART:20240419T160000Z DTEND:20240419T160000Z END:VEVENT END:VCALENDAR