The Science Data Enterprise: What Might the Future Look Like?

Big data. The Cloud. Citizen science. Teraflops. Exoscale computing. Ontologies and semantics. Linked data. These are some of the current buzzwords encountered in the context of contemporary science. Science data centers, such as the NASA Earth Observing System data centers, NOAA's geophysical data centers, and the Department of Energy's Atmospheric Radiation Measurement (ARM Archive) and Carbon Dioxide Information and Analysis Center (CDIAC), confront the challenge of determining the implications of these trends while still supporting their underlying mission. At the same time the core science is evolving and data centers are confronted with requests to demonstrate the relevance to domains such as decision support. How then to navigate these potentially conflicting demands? For individual data centers sources of guidance can be found from interactions with community-based groups such as the ESIP Federation. Other inputs come from science advisory groups and from program management. While the scale of these trends may have grown, exponentially in some cases, they are trends that have existed in the science data world for some time. This talk will examine these developments seeking to identify commonalities, challenges, and opportunities. More specifically, the presentation will be divided into three parts. The first part will seek to articulate some of the current trends and the implications of those trends for the science data enterprise. Second, the talk will seek to answer the question whether progress has been made on any of these dimensions. The final part of the talk will look forward through various perspectives, such as community-led groups and best practices, to present some future scenarios of where the science data enterprise is headed.