Cloud Infrastructure and Methodologies for GeoSciences: From Contrainers to Machine Learning and Artificial Intelligence

The availability of open access, petabyte-scale geophysical data creates new cross-domain analytical capabilities and challenges. Meeting the challenges of working with massive cross-domain data stores requires assessing existing methodologies and reworking them for an on-demand, distributed computing environment. This presentation examines existing data management and computing practices and introduces a framework for scientific cloud computing for the geosciences. Starting with cloud storage, the framework examines effective ways to leverage computing resources, including containers, serverless, and databases. In addition to addressing computing infrastructure, the framework also supports the use of computationally efficient software libraries that can parallelize workflows and leverage machine learning and artificial intelligence.