Entity-centric (people, places, things) data analysis requires both large volumes of detailed event data and more concise derivative data. Keeping data logically and physically close allows for high-performance analysis. Applications need to operate on the most relevant information, integrating batch and real-time calculations. Built on Apache HBase, WibiData can update records, continuously logging new information about users as they interact with a website, mobile application, product or service. The ability to add new ad-hoc columns to a table enables more flexible analysis. WibiData’s table layout management system, powered by Apache Avro, gracefully adapts to the changing data collection and analysis needs of your organization, seamlessly computing offline and online user data.
WibiData’s storage layer is built on top of Apache HBase. A three-dimensional storage model organizes information that would ordinarily require separate tables into a single condensed view that is sliced along a common characteristic. HBase allows for multiple columns to store data on a specific user’s interactions, over time. By allowing individual columns to be multivalued, content recommendations for a specific user can be modified based on the user’s interactions, creating a three-dimensional view of each user. HBase’s strengths lie in its flexibility: Rows can contain as many columns as you prefer with any names you choose. The “cells” in these columns can contain arbitrary bytes. This “schema-free” approach to data makes HBase a great platform on which to build a system.