A DISTRIBUTED ENHANCED SERVER FOR MULTIDIMENSIONAL SCIENTIFIC DATA

摘要

Faced with large and growing data sets from diverse sources (including general-purpose search engines that provide unwieldy results), the authors combine metadata search and data access concepts to create a prototype enhanced server tuned to scientific data search and accessibility. The dawn of the 21st century has seen an unprecedented amount of remote-sensing data collected. Around the world, national agencies have launched a series of Earth-observing and remote-sensing missions, all of which contribute to the amount of data available for scientific analysis. NASA's Earth Observing System satellite alone contributes more than half a Tbyte of data to various databases every day. In most cases, Earth science data comes in multidimensional arrays, but many variations in data format exist (such as map projections and spatial and temporal resolutions). Information volume and complexity are continuing to grow, together with requirements and usage stringencies imposed by different science communities in different environments. Using traditional data access and analysis methods, Earth science researchers will find it increasingly difficult to retrieve, analyze, and present results from such massive amounts of data. Scientists need distributed data information systems with effective search and access capabilities to perform their research more efficiently.