Hybrid BI Architecture Is Often the Best Approach
The definition of “a modern BI architecture” is elusive. It’s a slippery target, shifting with the ever evolving, complex landscape of analytics and its demands for better-faster-more data and better-faster-deeper insights.
This lack of definition doesn’t keep us up at night. At Senturus, we have long advocated that the “right” or “best” BI architecture depends on your organization’s current and long-term needs. The important thing – crucial, in fact – is that you don’t back yourself into a corner with an inflexible system design that does not adapt with your future needs. You want to lay the appropriate groundwork for flexibility.
Similarly, when asked about the value of an on-premise data warehouse versus a cloud-based platform, our answer is, “it depends.” More often than not, a hybrid architecture is the best approach.
To illustrate the point, we direct you to this whitepaper from IBM we recently ran across. While admittedly the paper has a vendor-biased agenda, it's still a good primer on the core benefits of hybrid BI architecture. It clearly illustrates how an approach that integrates traditional, on-prem sources (such as EDWs) with more modern, less-structured sources, (commonly sourced from the cloud), really makes a hybrid solution quite logical.