Design and Roadmap toward a NASA Earth Science Cloud

With the new high spatial, temporal, and spectral resolution capabilities in observing and computing systems, the pace of data growth is growing exponentially resulting in an urgent need for a new architecture for our data and computing systems. At the same time, new computational and information technology concepts have evolved to create virtualization technologies and made cloud computing a prevalent trend. How do we take advantage of these innovations as we are designing our next generation data and computing systems?

For many years, NASA's Earth science programs have invested in various efforts to modernize our computing, data processing, and data management systems. Some of the advanced prototypes and pilots such as NASA Earth Exchange (NEX) and Advanced Data Analytics Platform (ADAPT) in the past 10 years have demonstrated a successful new server-side data analytics workflow concept. In addition, the Earth Science Data and Information System (ESDIS) Project has recently embarked on building a major cloud-based framework for data ingest, archive, distribution, and management system called Cumulus. From these initiatives, a new Earth science cloud architecture is beginning to emerge.

We will present the design and roadmap to construct the NASA Earth Science Cloud. Building on the existing prototypes (e.g. NEX, ADAPT, and Cumulus), the new container technologies allow us to easily create a scalable science cloud. It may include the following steps:

- Convert, while architecting the solution for rapid prototyping of agile services, from existing prototypes to build and provide linkable, reusable microservices,

- package data, tools, workflows, models, and services into containerized objects,

- capture, manage and publish the analytics data processing and analytics provenance into an Earth science knowledge network,

- provide new search interfaces for any data, workload and service objects,

- integrate the data analytics platforms such as ADAPT and NEX with Cumulus, and

- enhance documentation, especially architecture, for future continuous development.