Visually exploring gigabyte data sets in real time

The article focuses on the virtual exploration of gigabyte data sets in real time. Where megabyte data sets were once considered large, data sets from individual simulations in the 300GB range are now found. But understanding the data resulting from high-end computations is a significant endeavor. Mathematician and pioneer computer scientist, Richard W. Hamming, pointed out that the purpose of computing is insight, not numbers. Analyzing large amounts of data presents a number of technical challenges. Simply getting all the data for analysis stresses even high-end hardware, it can take an hour to load a 100GB data set into memory. Loading the data little by little results in long times for a single pass through the data. Scientific visualization, or the use of computer graphics to represent data in ways that supports understanding, has played an increasingly important role in the analysis of large data sets. At NASA Ames Research Center's Numerical Aerospace Simulation Division, researchers are developing visualization systems for understanding very large data sets. INSET: Feature Extraction for Computational Fluid Dynamics.