Hybrid storage design for NC-94 database within the parametric data model framework

Abstract

The NC-94 dataset contains the most complete records of temporal and spatial variables for climate, crop, and soil in the north central region in the United States. Scientists store and process the dataset within scientific data formats which are efficient for scientific simulations. However, it is difficult for the public to access the dataset by using ad-hoc queries because the scientific data formats are not database management systems. This paper presents a hybrid storage model to build the NC-94 database which supports ad-hoc queries. The storage is designed within the parametric data model which has advantages over conventional relational databases. In our design paradigm, the database query language supports multirgranules as well as pattern windows to extract information, reducing the needs of high-level programming languages. We expect that our proposed paradigm significantly enhance the usefulness of the existing NC-94 dataset for the public domain. © Springer-Verlag Berlin Heidelberg 2006.