"Big Data as a Service..." Sounds like the ultimate buzzword mashup, doesn't it? Yet, there needs to be some ability to bring data resources into the enterprise, package it, and make it presentable and easily accessible to decision makers, without requiring them or IT to spend weeks and months finding, securing and cleaning data sources. Big data should be in the cloud — at least a private cloud.
Varun Sharma, an enterprise solutions architect, proposes a tiered big data reference architecture for capitalizing on the power and potential of data while ensuring security and governance. He notes how new organizations with innovative approaches — such as Tesla Motors — are relying on gobs of data from a range of devices and processes in the next generation of products.
Writing in Service Technology Magazine, Sharma provides some common-sense guidelines for making BDaaS a well-functioning platform within enterprises:
Ensure there is well-designed data governance. "Data governance is a must-have, and no longer merely a good-to-have," says Sharma. In today's extremely hyper-competitive markets, insightful knowledge means the difference between success and being overwhelmed. But it has to be based on the right data, based on business requirements.
Ensure data is protected. "Ignoring data security, data quality and data access can cost organizations millions of dollars, hurting enterprise agility, efficiency and reputation."
Pursue a tiered data strategy. "Break the operational tiers for data flow into logical groups," Sharma advises. The way to do this is by establishing a "consumption tier, analysis tier, organization tier and acquisition tier, to allow agility via loose coupling and abstraction."
Don't try to rush all data out to everyone all at once. "Consider the whole cycle from the acquisition of data to the extraction of information, and consider the hygiene factors along this path." There is a time in which data should be immediately available to decision makers, and there is a time when it can be retired.
Remember that BDaaS is everyone's puppy. "Successful Big Data-as-a-Service implementation would require close collaboration between Enterprise Architects, Data Architects, Database admin, BI and DW SMEs, SOA experts, InfoSec representatives and business strategists," Sharma writes.
An additional observation: What Sharma talks about here may not even be called "Big Data as a Service" two or three years from now — it may simply be data services. But it will certainly go a long way in supporting analytics-driven organizations.
(Thumbnail illustration: Joe McKendrick.)