<i>imMens</i>: Real‐time Visual Querying of Big Data

TLDR

Data analysts must interpret increasingly large datasets, sometimes containing billions of records. The paper presents methods for interactive visualization of big data that limit perceptual and interactive scalability to the chosen resolution rather than the number of records. The authors design scalable visual summaries using data‑reduction techniques such as binning or sampling, enable interactive querying across binned plots with multivariate data tiles and parallel query processing, and implement these in the browser‑based system imMens using WebGL on the GPU. Benchmarks demonstrate that imMens sustains 50 fps brushing and linking across dozens of visualizations, with performance invariant across data sizes from thousands to billions of records.

Abstract

Abstract Data analysts must make sense of increasingly large data sets, sometimes with billions or more records. We present methods for interactive visualization of big data, following the principle that perceptual and interactive scalability should be limited by the chosen resolution of the visualized data, not the number of records. We first describe a design space of scalable visual summaries that use data reduction methods (such as binned aggregation or sampling) to visualize a variety of data types. We then contribute methods for interactive querying (e.g., brushing &amp; linking) among binned plots through a combination of multivariate data tiles and parallel query processing. We implement our techniques in imMens, a browser‐based visual analysis system that uses WebGL for data processing and rendering on the GPU. In benchmarks imMens sustains 50 frames‐per‐second brushing &amp; linking among dozens of visualizations, with invariant performance on data sizes ranging from thousands to billions of records.