Changing BI Architectures and Approaches
In today’s fast paced society everything is trying to be bigger and better. We have grown accustomed to a society where everything needs to be more advanced and developed. This has resulted in an insatiable thirst for instant information. It is certainly no different in the business world and when we consider the rate at which technology innovation occurs today – where more and more data is collected, stored and analysed – the fast paced business realm is impacting the architectures of BI platforms and the approaches to BI delivery, and rightly so.
The business decision maker of today needs to have access to information instantly to be able to make the right decisions with a quick turn-around time. This requirement means that the BI industry is constantly undergoing continuous improvement around the delivery of BI and data warehousing, where the industry has noted that new BI architectures are needed as the BI world moves into what they are calling the second era of data warehousing. This era is called data virtualisation and alludes to the fact that the BI and data warehousing systems that IT have been building for over the last 15 or so years can now be considered outdated and do not have the facilities required to cope with new business requirements and the massive amounts of data a business typically deals with.
While I certainly agree with this industry sentiment, I believe that when looking at BI architectures, we, as an industry, must distinguish between two things – the one is the delivery delay, which refers to the time from requirement to report/dashboard/insight, and the other aspect is the problem of volume and structure of data, where the cloud and virtualisation fits into. While certainly both these concepts affect data and the role of BI, as experts in the BI industry, we need to distinguish between approach and architecture here and not get the two mixed up in the hype of terminology and trends associated with cloud and virtualisation.
You can have the biggest, most powerful, fastest BI platform and the most flexible and extendable architecture in the world (and that is exactly what is trying to be achieved with cloud-based storage and related capabilities – and successfully too), however, if the approach is slow and red-tape ridden, like the old “waterfall” type approaches, it is still going to take too long to get the data in the cloud and back out to the reports and dashboards. This is not due to the fact that the infrastructure is slow – on the contrary it is super fast – however, it is because the design and implementation process is slow and tedious.
Adaptable and extendable BI architectures, as well as reducing the analysis and implementation time (i.e. becoming more agile), are becoming more and more important in the overall BI industry, and that is what BI experts need to be paying attention to going forward. We need to seriously determine how technology innovation and the subsequent increase in data are impacting the way BI functions and map back to these changes effectively to ensure that BI remains the information technology that has a critical role to play in the business environment.
So to conclude, new technologies emerge regularly and with these new technologies more data and more opportunities are created – all of which impact the design and architecture of BI. As an industry, we need to always go back to the fundamentals of BI and remember its main purpose before merely blindly buying into the hype of a new industry trend. This will go a long way in keeping the credibility of BI going now and for years to come.
