Making Data Modeling a Thing of the Past

Share it on Twitter  
Share it on Facebook  
Share it on Linked in  

One of the things that hamper broad adoption of business intelligence is the need to model data in order to create a useful BI application. Obviously, a lot of business users don't have this skill and there are only so many folks in the IT department capable of supporting this task.

So it's with great interest that the BI community should be watching today the launch of Endeca Latitude 2, a business intelligence platform from Endeca Technologies. According to Endeca Chief Strategist Paul Sonderegger, it's now a whole lot easier to analyze data using Latitude 2 because regardless of its original source, data can be loaded directly into Endeca's MDEX Engine without first having to stage it.

The MDEX Engine is able to sidestep the need to model data by processing all data in memory on commodity servers. This means that instead of having to focus on creating a schema to perform the analysis, the actual data record is now the focus of the analytics.

With the release of the new version of Latitude, Sonderegger says that query performance is now 10 times faster, which means that Latitude can work against just about any source of Big Data by leveraging the inherent parallelism of multicore processors in order to analyze data in real time. In addition, Sonderegger notes Latitude 2 also now provides a semantic interface that makes it a lot easier for end users to access that data, while still giving IT managers the tools they need to manage it.



Sonderegger recognizes that in certain quarters of the BI community, this shift in focus will come across as heresy, but he says most software vendors have yet to fully appreciate the implications of widely available inexpensive memory. In contrast, Sonderegger says Endeca recruited Intel to become a minority investor just so that Endeca could gain early access to new in-memory processor technologies. Sonderegger says this capability is going to be critical to Endeca because applications today touch many sources of data. Trying to rationalize all that data into a common schema in order to do analytics is no longer practical, especially when customers want to know what is happening with transactions in real time.

Endeca today released InFront, a customer experience management platform that includes a new Infront Intelligence module for analyzing trends across multiple channels, including a variety of social media sources. This capability is important, notes Sonderegger, because e-commerce applications are now being embedded within social media networks.

With the rise of in-memory computing, something profound is begging to take shape in the world of analytics. Not only is more data than ever readily available with the rise of Big Data, that data is becoming more accessible because the cost of processing it is dropping exponentially. As that trend continues to evolve, it's starting to become clear that a major transformation in how IT organizations think about managing BI data is now well under way.