Data Analytics Infrastructure: Current Trends

Data analytics infrastructure is area that requires constant deep study to remain current.

The very term data analytics infrastructure is itself far from simple. It’s a wide ranging concept that comprises the many technologies and services that support the essential process of data mining for competitive insight. These many elements include managing, integrating, modeling and – perhaps most important – accessing the rapidly growing data sets that allow companies to better understand their business workflow and forecast market moves.

The challenge of data analytics is that it changes faster than you can say “business intelligence.” The technology itself is now undergoing rapid evolution, as is the techniques that practitioners are using. This is one sector where even an approach that has seen no refresh in a mere six months is already falling behind.

To provide a current snapshot, I’ll speak with Brian Wood, Research Director, Dresner Advisory Services. Wood will discuss the new report from Dresner, 2021 Analytical Data Infrastructure Market Study.

Among the questions we’ll discuss:

• What use case for ADI platforms did most respondents list as a top priority? What does this mean for the ADI market?
• It seem as if corporate standards have been a low priority for ADI, compared with security and performance. What changes do you think this trend will create?
• Is cloud or on-prem more popular for ADI platforms? What about the hybrid platform?
• Are there factors that make creating a coherent strategy for analytics projects difficult? (Like the range of innovation and the variety of ADI platforms.) How can business leaders deal with this challenge?
• Your sense of the future of the ADI market, several years out?

Listen to the podcast:

Watch the video:

Edited highlights from the full discussion – all quotes from Brian Wood:

Need for a Chief Data Officer

“One of the things that tends to help to limit this kind of [problematic] spread across the organization of different components is having a chief data officer, CDO or a chief analytics officer. Because that becomes a focus for them to make sure they have a cohesive and efficient analytic data infrastructure as opposed to a little of this here and a little of that there.

In most cases [the CDO] don’t really play the role of the cop trying to enforce it, although if you have the C in front of your title, you tend to get attention.”

Governance vs. Compliance

“To me, the only difference between governance and compliance is where the requirements come from. Governance is placing requirements on yourself. They’re internally. Compliance is from external.”

Shadow IT for Data Analytics

“Corporate standards [for data analytics practices] aren’t important. If one person finds a tool that is purely cloud-based and web-based and it works well for them, they will go ahead and buy it.

A lot of these tools and products have freemium models where someone can put their personal credit card in and use it for a month and then of course once they get used to the tool they’re not gonna let it go, and it becomes part of your analytic data infrastructure.”

Hybrid Reporting – and Hybrid Cloud

“One of the things that I find interesting is, even in the large organizations, they want everything in the Cloud, but they’re not starting from a green fields situation. They have lots of On-premise type of systems already.

But in order to get there from where you are today, you need a hybrid analytic infrastructure.

It has to report on the On-premise and the Cloud. But of course then you have multicloud as well. You have multiple public clouds, you have virtual private clouds. Having an infrastructure that will work with all of those, and I think particularly for the larger organizations that have been around for longer, it’s a stepping stone on the path if they wanna get to Cloud.

And most of them do. The survey says that is a preferred deployment approach for most industries and most functions. But in order to get there you have to go through the hybrid to get to a Cloud infrastructure.”

Future of Data Analytics

“So I’m often called an idealist, because I tend to look at the way things should be instead of the way they are. [chuckle]

So I’ll say, with a grain of salt, I will say that we will have AI capabilities that will enhance the way we do our jobs and not replace them. The future of work part aspect of it is one part of it, but realistically, we have models that do a lot of pieces of what a human brain does well, but there isn’t the master algorithm.

And so, what you’ll have is you’ll have the ability for an AI system to look at the different analytics in your organization and make recommendations, like, “This is good, but really you’re only using the trending. You’re not using the actual data.”

Obviously, now we’ve got models that beat the best chess players and all that. But you have to have something to model. So the way humans process information may or may not be the most efficient, but just taking that and putting it on a silicon substrate instead of soft wetware, so to speak – that helps a lot, but you still need the people to say, This is how I think. These are the connections that I make that led me to this conclusion.'”