Big data analytics 'needs segmentation' says Hitachi CTO

The future of big data lies not in in-house data crunching, however in outsourcing and off-the-shelf data analysis solutions, CTO of Hitachi Data Systems UK Francois Zimmermann has told Computing.

The big data debate

Zimmermann's comments came in response to Computing's recent coverage of the big data debate, in which Computacenter said that big data "hasn't happened but", and that a lot of stored analytics data is irrelevant.

"So I think what the market's doing is looking at some real, specific business use cases and saying it doesn't matter if it's user-generated, or social media, or machine-generated data; they've got a few values that they want to get out like that of analytics."

Zimmermann compared the growing big data industry to the so-called "cloud washing" phenomenon in the nascent cloud computing industry, when vendors were often accused of repackaging older products and services with a "cloud" spin in order to keep making sales.

Zimmermann said that companies who count big data as part of their business, just as Yahoo, can afford to be developers and to write their own usage cases. Nevertheless for others with operational and developer expenses such practice is "not very sustainable".

"For the companies where the only aim is business gain to put it more exactly than owning the innovation, I'd have thought they'd go down the packaged route. Companies will start to outsource it," said Zimmermann.

"As a matter of fact successful stuff has happened in big data when we've been doing business with the vertical data exactly," he added. "Focusing from a vertical perspective down – in other words than coming to it as a pure IT vendor up – is the way things are going to move. Analytics needs segmentation."

Join us for this Computing web seminar, in which the Head of BI at the Co-operative Group Nick Colebourn will be explaining just how he reigned in the Group's sprawling database estate and how significant savings were realised and data quality improved as a result.