Kognitio has launched a virtual cube builder, called Pablo, which is aimed at tackling businesses’ struggles around managing physical datasets.
Many IT departments spent a great deal of time managing hundreds or thousands of physical cubes, Kognitio said. Key challenges include the ever-expanding number of cubes to answer more and more database queries, and the latency associated with pulling in new data for up-to-date business intelligence.
The Pablo product, launched today and operating on Kognitio’s WX2 analytical database, offers the performance of physical cubes, but with the ability for users to build and query virtual cubes “on the fly”, it said.
The new system is also aimed at cutting costs by removing the need for businesses to run complex front-end applications. Data analysis can be carried out in a number of MDX tools, including Excel worksheets, ever popular with users.
It works with WX2 to create in-memory images of businesses’ transactional databases or data warehouses. Using a metadata layer and MDX connector, it takes data from a variety of areas – including CRM, financials and HR – into WX2 and can form virtual cubes as required on up-to-date information.
While some businesses were content with historic data from days or weeks before, many needed more up-to-date information in order to react quickly to emerging trends, it said. Examples of this could include retailers who want to capitalise on fast-moving customer buying trends and insurers who need to react to changes in risk and in the market.
The Data Warehouse Institute, which provides education on business intelligence and data warehousing, said the Pablo platform could help address the demands of complex data analytics for businesses.
Philip Russom, senior manager of research at TDWI, said businesses were looking for platforms “built for analytics” that support “in-memory databases, MPP architectures, and - of course - very, very large analytic datasets”. It was increasingly important, he said, that platforms quickly answered ad-hoc queries on complex datasets.