Time for service-oriented thinking in the big data space

'The sort of thinking that led to flexibly integrated SOA solutions should now be applied to data.'
Written by Joe McKendrick, Contributing Writer

Service oriented architecture thinking will help organizations manage the Big Data boom with the flexibility and agility needed for it to be valuable to the business.

That's the view of Steve Jones, a long-time proponent of SOA best practices and director of strategy for Big Data and analytics at Capgemini. In a new post. Jones urges that service oriented thinking be extended to data environments.

In spririt, SOA means doing away with rigid and monolithic approaches to systems, and breaking vital components down into digestable chunks of services. Jone proposes a "Business SOA" approach that will do away with the monolithic approaches to data management, in favor of more flexible, distributed approaches.

Data and associated analytics has for too long been structured as a "single canonical form, massive single schemas that aim to encompass everything." 

As Jones puts it:

"The sort of thinking that led to flexibly integrated SOA solutions should now be applied to data.  Get rid of that single schema, concentrate on having data served up in a way that matches the requirements of the business domains and concentrate governance on where its required to give global consistency and drive business collaboration.  That way you can ensure that the insights being created will be able to be managed in the same way as the operational systems."

In today's age, data analytics is no longer the highly latent "business intelligence" we've known for the past two decades. Instead of pulling archived data from data warehouses into BI tools to see what happened in the business over the last quarter, data needs to be available for analytics as soon as it's generated.  And it's not necessary to have everything standardized across the entire enterprise -- only those shared segments of data that matter.

In other words, data warehouses need to stop being "warehouses," and instead become transit points. 

And -- very importantly -- end-users should have the flexibility to build their own interfaces to access the information they need, without having to wait for IT to build it for them.  Is this not the spirit of SOA?

There's nothing new or radical about the idea of applying SOA principles to data, Jones reminds us. It's been proven to work, and makes sense that Big Data be available as services as well.

(Thumbnail: Joe McKendrick)

Editorial standards