Visualization of Databases: Detailed Visualization of Large, Complex Data

• Activity Number: 109
• Type: Invited
• Date/Time: Monday, August 5, 2013 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistical Computing
• Author(s): Ryan Paul Hafen*+
• Companies: Pacific Northwest National Laboratory
• Keywords: visualization ; large data ; R ; hadoop

Abstract

Visualization is a critical component of data analysis regardless of the size of the data, small or large and complex. Detailed data at the finest granularity must be visualized, not just summary statistics. In D&R we enable detailed visualization by Visualization Databases (VDBs), many displays, each an application of a visualization method to a statistically designed sample of subsets. Sampling can be representative, focused, or guided by cognostics: computer or human-generated diagnostics that distinguish attributes of interest for the visualization of each subset. An R package has been developed to allow users to rapidly prototype, create, and interactively view VDBs. The computing required for plot generation for large, complex data is achieved using the R and Hadoop Integrated Programming Environment (RHIPE). Plot viewing is facilitated through an interactive web interface to R.