| Abstract:
Viewpoints (vp) is a high-performance visualization and analysis tool for use with large, complex, multidimensional data sets, with dimensionalities that may surpass 100 and sample sizes that may exceed 10^6-10^8. Data sets of this size defy analysis with ordinary machine learning, exploratory data analysis, and/or visualization tools. Viewpoints was created for use with the extremely large data sets produced by current and future astronomical, earth science, and planetary missions, but it can be used for the analysis of large, multivariate data sets from any application. Viewpoints is designed to take extensive advantage of the hardware accelerated graphics (GPUs), multiple CPUs, and other capabilties of modern high-end desktop workstations. It is highly interactive, so that all feasible parameters are modifiable in real time with rich, immediate visual feedback. It's unique feature set high-performance, application-science driven development, scalability to multiple processors and the fact that it is specifically engineered for use with large datasets, sets it apart from existing exploratory data analysis tools.
|