Clustering Technology of a Data Engine for Analytical Computing

Contemporary scientific studies frequently rely on data-intensive analytical computing. While the main goal of this emerging form of computing is to facilitate hypothesis formulation or to test the validity of a postulated model, its primary method is usually that of data clustering. Since typical analytical tasks operate on very large volumes of potentially highdimensional data, scientific studies also face enormous problems of scale. This paper describes the clustering technology of a new engine for data-intense analytical computing. The technology is designed to operate in high-dimensional feature spaces without requiring dimensionality reduction. This enables the data engine to achieve high degrees of scalability and high interoperability between the analytical tasks. Most processes supported by the engine operate on a shared aggregate representation of data in the original feature space.