Most business users are already conditioned to explore data inside the context of an online analytics processing (OLAP) construct that presents them with a cube through which they can explore data. Starting today, Kyvos Insights wants to apply that same concept at scale to Big Data.
Emerging from stealth today, Ajay Anand, vice president of product management for Kyvos Insights, says the company has developed OLAP software that enables end users to create and analyze cubes of data that reside directly on Hadoop. Rather than having to move specific sets of data into a data warehouse that provides OLAP functionality, Anand says end users can now directly manipulate Big Data inside the repository where it is now most naturally stored.
Compatible with the major distributions of Hadoop, Anand says the goal is to make it simpler to analyze a wide variety of data types within a familiar multi-dimensional OLAP construct. End users can then make use of visualization tools from Tableau Software to further explore that data at their leisure, says Anand.
While Kyvos Insights is not the first startup company trying to marry OLAP cubes to Hadoop, Anand says Kyvos Insights is making use of a massively scalable OLAP engine to process Big Data at an unprecedented level of scale.
Of course, as interesting as that may be to the average business analyst, the implications of being able to run OLAP cubes on Hadoop are nothing short of profound. The primary reason most IT organizations move data into a data warehouse is to move into an OLAP cube for analysis. If that data can now be just as easily accessed on Hadoop, it brings into question the whole value proposition of building a data warehouse in the first place.