Subscribe

BI: Architecting new solutions in complex environments

By Tracy Burrows, ITWeb contributor.
Johannesburg, 19 Feb 2013

ITWeb Business Intelligence Summit and Awards 2013

The 8th Annual ITWeb Business Intelligence Summit and Awards takes place on 26 and 27 February 2013, with a workshop on 28 February. Themed "Integrated BI for optimised performance", the 2013 summit empowers BI practitioners to derive the maximum value from their BI implementations. For more information and to reserve your seat, click here.

Delivering the increased insight promised by the latest wave of technologies for business intelligence (BI) and big data requires taking a look at the big picture first.

This is according to Kevin Lancaster, big data and business analytics director for Engineered Systems at Oracle EMEA.

Lancaster says BI is getting more complex, with advances in big data, mobile BI and data visualisation tools.

At the same time, modern BI is operationally significant, which has further implications for the technology and architectural choices of IT. For example, data warehouses are no longer designed only for a small number of analysts. Today, they must also support a highly varied workload, large numbers of concurrent users, and real-time data updates to underpin operational systems that depend on the analytics.

"The demands for big data analysis and mobile BI, plus the popularity of in-memory processing and highly interactive data visualisation tools, are bombarding IT teams at a time when they have less budget rather than more, and while business requires greater agility than ever before," says Lancaster.

"The challenge is not the availability of all these technologies, but figuring out how to bring them together efficiently, with minimal disruption to the business, and in an easy-to-manage, cost-effective way."

He says Oracle is tackling this problem from two directions: bottom-up, by combining technologies into building blocks that are pre-integrated, pre-configured and pre-installed, and optimising them for specific parts of the solution to make it easier and faster to deploy and less costly to manage; and top-down, by helping its customers develop a blueprint or overall solution architecture based on capability and the business and technical requirements to guide their information management efforts.

Lancaster says that while businesses such as Oracle are working on solutions that help address these challenges, there is no single 'killer technology' that can do it all.

"Different technologies have different strengths - you need a blend. Oracle's strategy is to simplify IT and it is being delivered through a combination of huge R&D investment, strategic acquisitions and partnerships."

Another aspect is the need to implement new capabilities within the context of the existing IT landscape, says Lancaster.

"If you don't have an overall design or blueprint, introducing new technologies can actually make the problem worse - you can unintentionally introduce new information silos and IT complexity.

"Without a clear vision of where you are headed, it's very hard to get there" he notes.

Lancaster says organisations need to design and architect solutions that can withstand change, handle growth, incorporate new technologies as they become relevant, and are cost-effective.

By working towards a 'big picture' and simplifying the IT landscape, IT can focus on enabling business growth and innovation, instead of just 'keeping the lights on'.

He says: "A top priority for IT of course is to keep everything running to an agreed level of service. But this is not driving the business forward. We need to do those things faster, at scale, with more detail, and be much more agile.

This is where IT delivers real value. Rebalancing the IT budget by spending less on day-to-day 'business-as-usual' and more on enabling the business to innovate is the goal."

For example, he says: "It was once common for IT to invest a year or more in planning, procuring, assembling, testing and then tuning new hardware platforms from multiple components from multiple vendors for an enterprise data warehouse, and to repeat this every three years or so as the hardware was refreshed.

"This puts tremendous strain on highly skilled people, who could otherwise be doing something truly innovative for their business. Now, the default approach is to purchase highly tuned, optimised appliances, which can be installed in days, cutting months off the process. Oracle has taken this concept, now de-rigueur for the traditional data warehouse, and extended it across the whole IT landscape," he says.

Lancaster will speak at the upcoming ITWeb BI Summit on "Engineered Systems for Business Analytics, and the Reference Architecture for Information Management". For more information on this event, click here.

Share