This morning, Datameer released version 4.0 of its self-titled analytics and now integration platform for Hadoop. The company has brought its visual style of analytics to the entire data integration process, allowing users to see data normalization changes immediately.

Karen Hsu, senior director of product marketing at Datameer, said that version 4.0 provides visual feedback from a Hadoop cluster’s data stores. This allows those mining the data to “see things like completeness, accuracy, and if there are duplicates. These are things traditional architects have done for years in the business intelligence and integration scene. What we’re enabling with 4.0 is this ability to see the shape of the data, and for an analyst to adjust it without having to rely on IT,” she said.

(Related: Other Big Data news, this time out of CeBIT)

Datameer 4.0 gives immediate visual feedback on data manipulation, said Hsu. “The challenge analysts and data scientists are facing at every step of the process is that people will spend 80% of their time doing data preparation, and joke that they spend the other 20% of the time complaining about it. Data preparation is a big issue now.”

The standard spreadsheet-style view of Hadoop data can be flipped over in Datameer 4.0 to show visual representations of the data being manipulated. This new feature (or view) is called “Flip side” by Datameer, and it can be invoked at any stage in the process of preparing for or building analytics.

“Many other ingestion vendors don’t have that ability to make a change and immediately see those results,” said Hsu.
“With Big Data, because you are bringing in so many different sources of data, it’s paramount you be able to see how the data looks after you bring it together, and bringing it together is not just a once-in-time process, it’s something you do continuously.”

Datameer 4.0 is available today, and an introductory webinar will be held by Datameer on Thursday.