Click on the screenshot to see a flash demo of a simple BASIN session.
Beowulf PC clusters are now widely used in many branches of science where they have become key tools for the analysis of large datasets resulting from laboratory experiments, field observations or numerical simulations. However, although Beowulf clusters have been of great benefit to scientific research, no publicly available parallelized software package exist to perform even the most common operations needed in the analysis and visualization of large datasets. In sharp contrast with the situation for non-parallelized data analysis, where a wealth of software packages performing standard data-reduction tasks are available to researchers (e.g. IRAF, MIDAS), for parallel data analysis and visualization each individual group has tended to develop its own set of specialized tools. The increasing diffusion of multi-processor PCs further increase the demand of software packages for parallel data analysis.
BASIN is a flexible, integrated suite of tools for parallel data analysis and visualization which has been designed and is currently been developed to address these concerns; BASIN is aimed at providing researchers with a package which allows them to harness the power of Beowulf PC clusters and multi-processor machines without necessarily being experts in parallel programming. While BASIN is motivated by problems particular to astrophysics, its core functionalities and parallel data distribution and manipulation are general end provide a number of general to easily develop libraries specific to data analysis in a number of different fields.