SOURCE: Zoomdata

February 19, 2015 07:01 ET

Zoomdata Adds New Spark DataFrames Engine for High-Scale BI

New SparkIT Feature Delivers Fast Visualization of Legacy Database and Flat File Data

SAN FRANCISCO, CA--(Marketwired - Feb 19, 2015) - Zoomdata, developers of the Zoomdata big data analytics and visualization system, today announced a new engine in the latest release (version 1.5) that integrates Apache Spark and Spark DataFrame technology natively into its architecture. Where other BI tools use a proprietary data server layer (often Windows-based), Zoomdata now leverages Apache Spark and Spark DataFrame technology for all post-query data calculation, aggregation, pivoting, filtering, and result set caching. This allows Zoomdata to use the resources of Spark clusters to perform BI and data visualization at extremely high speeds and scales, while being data-level interoperable with other Spark technologies and processes. Zoomdata version 1.5 is available as a Linux download, Docker container, and live demo at

Zoomdata 1.5 also adds Zoomdata SparkIT technology, which enables speed-of-thought analytics on data in S3, HDFS, or legacy Oracle, SQL Server, Postgres, and MySQL datastores, by intelligently calculating and storing cubes of data as interoperable Spark DataFrames in a Spark cluster. Once in Spark, Zoomdata enables lightning-fast analysis and visualization. Zoomdata installs with an embedded Spark instance suitable for getting started, and can then be pointed to a larger external Spark cluster for theoretically infinite data and user scalability.

The new Zoomdata Spark DataFrame engine can be used with Zoomdata's patented data sharpening technology with data in modern stores such as Cloudera Impala, Cloudera Search, Elastic Search, SparkSQL, MongoDB, SolR, Redshift, and Aurora. Together, DataFrames and data sharpening allow for data to sharpen in front of a user's eyes while it is being initially analyzed by a modern big data back-end. Further operations on the results such as filtering, sorting, pivoting, and changing graph styles are then performed instantly from the resulting DataFrames using Spark.

By combining SparkIT, data sharpening, and Spark DataFrames, Zoomdata provides the world's fastest analytical and visualization experience on data sourced from modern big data backends, flat files in S3 and HDFS, as well as slower legacy databases and data warehouses. Zoomdata gives the user a view of the results from a single pane of glass at extremely high levels of data and user scale by leveraging an external Spark cluster.

"Zoomdata has been architected from the ground up to natively leverage the latest big data technologies on the back-end, while proving a beautiful design-first ease of use experience on the front-end," said Justin Langseth, CEO of Zoomdata. "Apache Spark has allowed us to leverage massive external Spark clusters to perform data operations in an interoperable, scalable way, instead of building our own proprietary data crunching and storage layers as most of our competitors have done in the past. By embracing Spark and Spark DataFrames we are able to level the playing field between next generation big data backends, flat files, and legacy databases, and provide consistently great user experience regardless of the back-end data store."

Zoomdata is also releasing its next-generation pivot table visualization, based on the new Spark DataFrame engine, which can provide fast interactive pivoting and analysis against billions of rows of data. The Zoomdata pivot table can report on thousands of columns, and an infinite number of rows, all from within a web browser.

Zoomdata 1.5 adds other new features that improve performance, ease of use, visualization flexibility and more, including:

  • Visualizations: New out-of-the-box visualizations including a World Map, Stacked Bars, Donut and an updated version of the Heatmap;
  • Connectors: A new connector for SparkSQL; plus support for Flat JSON, Flat XML and TSV files;
  • Tutorials: New tutorials in the Help section make it easier for customers to quickly get started with Zoomdata.

More and more Zoomdata customers have discovered the power of its patented Data Sharpening architecture. Instead of waiting minutes or hours for a query to run to completion, Zoomdata uses unique data sharpening technology to provide an instant approximated sketch of the results. Then it continuously updates the visualization through stream processing of additional micro-query results until the data fully sharpens. Because Zoomdata has a stream-processing engine at its core, it can also seamlessly analyze real-time as well as historical data. And due to Zoomdata's modern HTML5 and touch-first interface, users can explore billions of records in seconds not only through web browsers, but also via mobile smartphone and tablet touch devices.

In addition, Zoomdata gives developers access to a JavaScript SDK through which they can quickly and easily build a private-labeled bespoke experience that leverages big data (

About Zoomdata, Inc.
Zoomdata, developers of the world's fastest big data exploration, visualization and analytics platform, lets business users see and interact with data in all new ways. Designed mobile and touch first, its patented micro-query architecture delivers results on billions of records in seconds and gives users a single plane of access for bridging old data and new data. Zoomdata is backed by Accel Partners, NEA, and Columbus Nova Technology Partners. For more information, visit: