Open NASA Earth Exchange (OpenNEX): Strategies for enabling cross organization collaboration in the earth sciences

Thursday, 18 December 2014
Andrew Michaelis1, Sangram Ganguly2, Ramakrishna R Nemani2, Petr Votava3, Weile Wang4, Tsengdar J Lee5 and Jennifer L Dungan2, (1)Organization Not Listed, Washington, DC, United States, (2)NASA Ames Research Center, Moffett Field, CA, United States, (3)California State University Monterey Bay, Seaside, CA, United States, (4)CSUMB & NASA/AMES, Seaside, CA, United States, (5)NASA, Burke, VA, United States
Sharing community-valued codes, intermediary datasets and results from individual efforts with others that are not in a direct funded collaboration can be a challenge. Cross organization collaboration is often impeded due to infrastructure security constraints, rigid financial controls, bureaucracy, and workforce nationalities, etc., which can force groups to work in a segmented fashion and/or through awkward and suboptimal web services. We show how a focused community may come together, share modeling and analysis codes, computing configurations, scientific results, knowledge and expertise on a public cloud platform; diverse groups of researchers working together at “arms length”.

Through the OpenNEX experimental workshop, users can view short technical “how-to” videos and explore encapsulated working environment. Workshop participants can easily instantiate Amazon Machine Images (AMI) or launch full cluster and data processing configurations within minutes. Enabling users to instantiate computing environments from configuration templates on large public cloud infrastructures, such as Amazon Web Services, may provide a mechanism for groups to easily use each others work and collaborate indirectly. Moreover, using the public cloud for this workshop allowed a single group to host a large read only data archive, making datasets of interest to the community widely available on the public cloud, enabling other groups to directly connect to the data and reduce the costs of the collaborative work by freeing other individual groups from redundantly retrieving, integrating or financing the storage of the datasets of interest.