STATETECH: You previously said, “...We build a foundational bottoms-up infrastructure that enables analytics and data-driven decision-making.” What does this mean for agencies?
REIN: New Jersey has a 2023-2025 strategic plan. It’s neither an operational nor a tactical plan, but rather a directional plan. Toward that end, it’s increasingly important to use data as we provide government services. We have data, and we need, therefore, analytics as well to inform our cabinet members, our governor, our chief of staff, our residents and our policymakers.
We have many agencies, and they are quite diverse. In New Jersey — and this is probably true everywhere — if you’ve seen one agency, you’ve seen one agency. They’re all different; so is their technology maturity, their technology budget and their staff maturity. So, to your question, I don’t know that it strictly says we keep our data here and our applications there. It’s not that, strictly speaking. I don’t think it’s that well-defined, and it’s certainly not a requirement to architect applications that way. For some data, it makes operational sense to have the data coexist with the application. But for some, in many cases, it does not.
For example, like many states, we still have a mainframe. The mainframe, by and large, is 60 to 80 percent of a given entity’s processing when it comes to large back-end processing. It’s processing all of the motor vehicle licenses, all the unemployment claims and more. Now, we have moved our mainframe into the cloud through Mainframe as a Service. Our mainframe data is not local, but when we’re providing Mainframe as a Service, it is available to all those other applications as though it were on-premises.
STATETECH: So, you provide customer service to your fellow agencies. They come to you with a technology need. And you provide them with what may be a hybrid cloud solution.
REIN: Of the many hundreds or thousands of applications, it’s probably an extremely small number that are stand-alone applications. There are a few, but they’re rare. Therefore, the implication is that almost every application must interface with or send update statuses to other applications, and those applications could be on-premises or in the cloud. The Office of Information Technology’s job is to provide the infrastructure. The agencies own their applications.
So that means the business intelligence — how this application works, what data’s in our database, all of that — is, in the model that we operate within, owned by the agency. So they have programmers, but they typically don’t have server and network professionals. That’s us. So, this is how we draw the distinction of responsibilities. It is part of my job to approve every technology-related procurement.
There’s not a clean way of making these decisions, but there are several factors to consider when making the decision of where to host a given application. We may grow something in AWS or grow it in Azure, and it may depend on what other applications this thing has to talk to. We may look at other factors if we are going to be on-premises.