Incrementally Exploring Large-scale Multidimensional Data
Jaemin Jo, Wonjae Kim, Seunghoon Yoo, Bohyoung Kim, and Jinwook Seo / 2017
The advance in distributed computing technologies opens up new possibilities of data exploration even for datasets with a few billion entries. In this paper, we present SwiftTuna, an interactive system that brings in modern cluster computing technologies (i.e., in-memory computing) to InfoVis, allowing rapid and incremental exploration of large-scale multidimensional data without building precomputed data structures (e.g., data cubes). Our performance evaluation demonstrates that SwiftTuna enables data exploration of a real-world dataset with four billion records while preserving the latency between incremental responses within a few seconds.