Use Case

A Cloud-Based Framework for Visualization and Analysis of Big Geospatial Data

V. Chandola (University at Buffalo) with R. Vatsavai (North Carolina State University), P. Hogan (NASA Ames Research Center), B. B. Bhaduri (Oak Ridge National Laboratory)

What?

• Integration, visualization, and analysis of geospatial data from around the world

Why Aristotle?

• Offloading of variable computational demands from local infrastructure
• Expert service hub for widely distributed data
• Scalable and elastic computing

Accomplishments

• Developed the capabilities to create Spark analytical clusters on-demand.
• Deployed a new distributed machine learning method for change detection in sustainability data.
• Ran scalability tests on Red Cloud at Cornell before migrating the application to Lake Effect cloud at the University at Buffalo.
• Developed webGlobe, a browser-based, cloud-driven interactive 3D user interface that allows scientists to upload, visualize, and analyze Network Common Data Form (NetCDF) data sets and is, at present, the only browser-based system available with this functionality.
• Completed initial runs of Gaussian Process-based change detection algorithm on 200 years of climate simulation data.
• Refined webGlobe’s integrated visualization and analysis capabilities that support a variety of data formats prevalent in the climate and earth science communities.
• Continued development of an Energy-Water Knowledge Discovery Framework portal using webGlobe technology currently running on the Aristotle cloud.
• Performed comparative evaluations of various machine learning methods on Aristotle resources to better understand the Energy-Water nexus.
• Shipped a version of webGlobe to collaborators at Oak Ridge National Laboratory to support their research activities in the area of climate data analytics.
• Co-organized panel and gave poster presentation at the AGU Meeting on "Application of Information and Data Science Methods and Technologies to Climate Research and Energy-Water Knowledge Discovery."
• Developing new machine learning components to enhance webGlobe and a Jupyter Notebook extension of the framework to allow users to analyze underlying data.
• Created an interaction platform and tailored webGlobe for the Lower Great Lakes Resiliency Effort. Both are hosted on Aristotle.

Plans

• Continue development of iGlobe-based application to support sustainability research. Plan to create a data and computing infrastructure, powered by Aristotle, to support a large collection of researchers, community stakeholders, and planners, working on the sustainable development of the Lower Great Lakes area.
• Currently working on a paper that describes the OUTSTEP Community Platform and its digital capabilities.

Products

• Technology demonstration at the NSF-funded Organizing Urban Transects for a Sustainable Transformation of Economic Partnerships across the Lower Great Lakes Workshop (OUTSTEP 2019).
• Zaidi, S.M.A., Chandola, V., Allen, M.R. & Bhaduri, B.L. (2018). WebGlobe - A cloud-based geospatial analysis framework for interacting with climate data. Proceedings of the 7th ACM SIGSPATIAL International Workshop on Analytics for Big Geospatial Data.
• Zaidi, S.M.A., Chandola, V., Allen, M.R. & Bahduri, B.L. (2018). Anomaly detection in energy-water nexus. Poster at the American Geophysical Union Fall Meeting, Washington, DC and co-organizer of a panel on the Application of information and data science methods and technologies to climate research and energy-water knowledge discovery.
• Mohammed, S.Z.A., Allen, M.R., Chandola, V. & Bhaduri, B.L. (2018). Machine learning for the energy-water nexus: challenges and opportunities. Big Earth Data Journal.
• Sharma, A., Mohammed, S.Z.A. & Chandola, V. (2018). A cloud-based geospatial analysis framework for interacting with climate data. Proceedings of the 26th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems.
• Allen, M., Mohammed, S.Z.A., Chandola, V., Morton, A., Brelsford, C.M., McManamay, R.A., Binita, K.C., Sanyal, J., Stewart, R.N. & Bhaduri, B.L. (2018). A survey of analytical methods for inclusion in a new energy-water nexus knowledge discovery framework. Big Earth Data Journal.
• Morton, A., Piburn, J., Stewart, R. & Chandola, V. (2017). Development of a suite of analytical tools for energy and water infrastructure knowledge discovery. Presentation at the American Geophysical Union Fall Meeting, New Orleans, LA.
• Sanyal, J., Chandola, V., Sorokine, A., Allen, M.R., Berres, A., Pang, H., Karthik, R., Nugent, P., McManamay, R. & Bhaduri, B.L. (2017). Towards a web-enabled geovisualization and analytics platform for the energy and water nexus. Presentation at the American Geophysical Union Fall Meeting, New Orleans, LA.
• Morton, A., Piburn, J., Stewart, R. & Chandola, V. (2017). Leveraging advances in population modeling to support energy and water nexus knowledge discovery. Poster at the American Geophysical Union Fall Meeting, New Orleans, LA.
• Co-Chaired 6th ACM SIGSPATIAL International Workshop on Analytics for Big Geospatial Data (BigSpatial-2017).
• WebGlobe Capabilities video (2017) demonstrates webGlobe, a new tool developed by University at Buffalo use case scientist Varun Chandola. Powered by Aristotle, webGlobe provides access to and visualization of remotely available NetCDF geo data.
• Tran, D., Khoo, E.R. & Chandola, V. (2016). WebGlobe: Big geospatial analytics on the cloud. Presentation at MITRE Corporation, McLean, VA.