Skip to main content Skip to secondary navigation
Journal Article

Demonstration: MacroBase, A Fast Data Analysis Engine

Abstract

Data volumes are rising at an increasing rate, stressing the limits of human attention. Current techniques for prioritizing user attention in this fast data are characterized by either cumbersome, ad-hoc analysis pipelines comprised of a diverse set of analytics tools, or brittle, static rule-based engines. To address this gap, we have developed MacroBase, a fast data analytics engine that acts as a search engine over fast data streams. MacroBase provides a set of highly-optimized, modular operators for streaming feature transformation, classification, and explanation. Users can leverage these optimized operators to construct efficient pipelines tailored for their use case. In this demonstration, SIGMOD attendees will have the opportunity to interactively answer and refine queries using MacroBase and discover the potential benefits of an advanced engine for prioritizing attention in high-volume, real-world data streams.

Project page

MacroBase is a new analytic monitoring engine designed to prioritize human attention in large-scale datasets and data streams.
Author(s)
Peter Bailis
Edward Gan
Kexin Rong
Sahaana Suri
Journal Name
SIGMOD ’17: Proceedings of the 2017 ACM International Conference on Management of Data
Publication Date
May, 2017
DOI
10.1145/3035918.3056446
Publisher
ACM