An InetSoft BI Webinar: Aligning the Right BI Technologies to the Business Needs

This is the continuation of the transcript of a Webinar hosted by InetSoft on the topic of "Agile BI: How Data Virtualization and Data Mashup Help". The speaker is Mark Flaherty, CMO at InetSoft.

Mark Flaherty (MF): In aligning the right BI technologies to the business needs, it’s important to understand change, how to not just deal with it, but embrace it. Be able to adapt to those things faster than the competition and likewise take complexity and make it a friend. I mean use that agility to provide competitive advantage but tame it down to the point where people can actually deal with it.

With that kind of a focus on agility, what’s happening again in enterprise architecture is there is sort of a shift away from the classic boxes application, infrastructure, information, data architecture. Fundamentally there is a more of a business capability oriented thinking that might actually live in the business organization.

You may call it business architecture because it answers the questions, what do we need, what should we be able to do? They may come from product management kinds of folks, strategic planning kinds of folks etc. who then drive the creation of solutions based on their knowledge of what technology should be able to do. In this scenario, IT focuses on how to deliver that BI solution based on the expertise that they have on the different BI systems available.

In this model the role of information and data is pretty critical. They not only play the role of being able to connect those two layers, broad layers, if you will, but also they act as the way to link information up and down. Obviously the business logic and processes are embedded there. The resulting information agility has to provide some advantage. So the foundation of the business capability to an architecture is flexible, virtualized access to diverse information sources.

Now we will be exploring what data virtualization means. Virtualization has a couple of different connotations, abstraction connotations, the connotation of providing data when asked for, not necessarily in real time, but close to that, or in right time, if you will. And then the types of information sources tend to be pretty diverse, not just what's in the data center, not just what's in content management systems, or desktops, but also external partner data, big data, cloud-based data, social media, and even unstructured information.

The challenges in this context are not new. The data integration challenges are similar to what they would always be, except that they are growing in complexity, growing in magnitude. There’s really an exponential growth in data that is being collected and saved. We will see some quotes here shortly on what's happening the data silos. It isn’t just within the organization but also the customer voice coming from the Internet in the form of unstructured data, big data coming from the cloud, partner information, log files, etc. All of these add complexity, add more rigidity if you haven't started to make them flexible, and they can add high costs in maintaining them.

One statistic I have come across is that the quantity of data is growing three and a half times faster than the available storage, even though storage is getting cheaper. And imagine if you on average replicated that data 3 to 7 times, you have a lot of information that you have nowhere to put. So this idea of virtualization is basically gaining currency very rapidly, and as I mentioned earlier it plays multiple roles in the enterprise architecture.

Learn about the top 10 features of embedded business intelligence.

This is a simplified picture of an enterprise architecture. It’s meant to be simplified, but the first idea is that you want to virtualise access to disparate sets of data sources. What that means is abstraction, decoupling, representing different data structures, data types, data connections, data sources as a logical view if you want to call it by using a data mashup tool to connect them. The term logical view is a limited term but effectively it's an intermediary layer on which you can build other views of your data. It is not to be construed in the narrow sense of a database view, but that’s really what virtualized access to data sources mean.