Operational BI for collaborative analytics using notebooks

Summary

The world of BI has long relied on rollups, nightly summaries, canned queries and reports on historical data to provide insights to users. As data volumes grow, the need to collaboratively define data analytics has led to the growth of notebook style visualization aids like Jupyter, Apache Zeppelin etc. The fundamental idea here is that notebooks can work directly with live data sources (like Spark/SnappyData) and allow users to define adhoc queries, the results of which can be visualized in the notebook. Furthermore, the output from these notebooks can be easily shared for building out collaborative analytics in real time, making Operational BI a much more dynamic capability within the enterprise. From retail to manufacturing to finance, operational BI improves business agility and decision making by making real time information available to decision makers at the right time

Challenge

  • Working with a large number of legacy data sources
  • Combining legacy data with new incoming data
  • Defining new queries and reports dynamically and collaboratively
  • Dealing with different languages, data storage protocols and system incompatibilities in general

Solution

Synopsis Data Engine from SnappyData is a revolutionary new technology that combines SnappyData’s data unification capabilities with mathematically proven statistical techniques to provide highly accurate answers to aggregate class queries in real time orders of magnitude faster.

Snappy Capabilities in Use

  • Synopsis Data Engine
  • Data unification using SnappyData
  • Apache Zeppelin integration for instant visualization

The Apache Spark Database

SnappyData is Spark 2.x compatible and open source. Download now