Flushing Out a Big Data Outsourcing Offering May be Years Away

By Dan Berthiaume

Big Data is not yet having the impact on outsourcing that other leading edge technologies such as cloud, mobile and social are creating, but its influence is growing. “It’s the least mature of the ‘Big Four’ emerging technologies so it’s had the least input on the broader sourcing industry,” said Stanton Jones, emerging technology analyst at sourcing advisory firm ISG.

Despite the fact Big Data is “quite immature” as a business technology, Jones expects its impact to grow significantly in the future. “The key Big Data players are open source technologies such as Hadoop with Apache,” he said. “It usually takes a long time for IT shops to commoditize and commercialize open source offerings.”

As a result of the relative scarcity of commercialized Big Data packages, Jones said most Big Data implementations are “amalgamations” of hardware and software, with most of the software of the open source variety. Jones said the large organizations that form the bulk of ISG’s client base are interested in something more substantial.

“A lot of our clients are talking to traditional Tier I and Tier II IT service providers, as well as the big Indian service providers,” he said. “There are few mature Big Data offerings from the big service providers.”

Service Providers Eye Big Data

But this doesn’t mean major IT service providers aren’t actively trying to develop Big Data solutions. “They are selling a combination of hardware and cloud services to form the infrastructure for Big Data,” Jones said. He added many IT service providers are also offering analytic services and software to form a middle layer supporting Big Data operations, but there is no classic “turnkey” Big Data solution in the current outsourcing marketplace.

“Nobody has fleshed out the Big Data stack into an offering,” Jones said. “That is years away.”

In addition to its strong reliance on open source software, another aspect of Big Data that makes it difficult to create a turnkey outsourced solution is its decentralized nature. “The underlying infrastructure is based on highly virtualized distributed horizontal information platforms,” said Jones. “Big Data is not centralized with IT. It’s flatter and more horizontal. It moves with computing power rather than sit in a central repository like a data warehouse.”

Securing Big Data

Big Data security also poses some issues to outsourcing clients. “Big Data democratizes IT and more people than just those in the IT department use the information layer that enables Big Data,” said Jones. He said Big Data is so new that standard security policies are “yet to be determined.” The fact that Big Data usually consists of a collection of disparate hardware and software and is represented by a single internal champion rather than a department also complicates efforts to create security and governance policies.

Thus, Jones said Big Data security and governance policies are usually created in real time as Big Data systems are developed. “It’s not an ideal way to do it but Big Data is moving so fast that nobody has thought about security until it’s sitting on their desk.”

Service Providers & Big Data

Jones mentioned one final sign that Big Data has a definite future in outsourcing – he believes that service providers are likely dabbling in Big Data for internal uses just like their clients are. “The general internal trend is incubatory – slowly start building it out,” he said.

Service providers are most likely not far ahead of their clients in the use of Big Data, Jones concluded. “They’re like cobblers’ children,” he said. “They’re good at building Big Data solutions for clients but not quite so good at building them for themselves.”