This content is provided by Splunk.

Data is increasingly the backbone of decision making, which is why the Defense Department is putting so much effort into ensuring a secure, unhindered flow of data between warfighters and leadership with programs like Joint All-Domain Command and Control (JADC2). But bringing the entire DoD enterprise under a single unified data environment is a huge undertaking fraught with obstacles. So how can DoD build its future data enterprise?

Challenges

For one thing, DoD is going to have to figure out how to secure vulnerable data at the edge.

“Nothing really matters if you can’t connect. Warfighters find communications totally cut off in delayed or disconnected, intermittent low bandwidth environments,” said Kate Mercer, a Booz Allen vice president and leader of the firm’s Next Generation Data Solutions team within the Defense digital business. “We also see adversaries deploying increasingly sophisticated tactics, where they’re trying to capture information, whether it’s at rest, in transit or on devices traveling with our military teams. I really think we are only as good as our tactical edge. It sets the bar for our success.”

Another challenge is the way data is currently stovepiped. DoD has invested in several different data platforms, making a great deal of data unsearchable, undiscoverable by analytic tools. Sometimes different data architectures and classification levels exist on the same network.

And then there’s the sheer volume. The amount of data produced every day continues to increase exponentially, as does the number of sensors gathering said data. That’s lead to some data systems that are simply overwhelmed, and unable to identify and analyze useful data.

Interoperability

Having an overall data strategy is important, Mercer said, but DoD should also be looking at its reference architecture. That should be guided by open systems architecture in order to foster interoperability among the various data platforms.

“Having those standards that each of the data platforms can orchestrate and implement within their own data platform so that we can have that connective tissue across the enterprise, I think that that’s something the DoD is really leaning on private industry to bring to the table,” Mercer said. “Having that open systems architecture approach is critical for being able to connect all the existing investments that they have across the data platforms, in order to facilitate data sharing across the enterprise.”

There’s also a culture aspect to this.

“Operators would say that they have always functioned as a joint domain. However, in reality, each service brings its own language and standards,” Mercer said. “As we bring data together in those common standards, that’s going to increase the capacity, the productivity, the capabilities of our operators, to have not only those common standards, but training. And it just really eases the barriers in the joint domain.”

Securing the data

The more endpoints you have, the greater your cyber risk. Especially in DoD’s unique circumstances, where endpoints could potentially end up in the hands of adversaries. That’s why it’s important for DoD to have multiple layers of access control. A Zero Trust model allows DoD to verify all devices at the edge. Edge data storage allows DoD to break up storage locations, eliminating a single access point. And decentralizing the network through multiple edge locations reduces the ability for denial of service attacks.

“We’re familiar with encryption, endpoints are plentiful, but if we can include the latest and bleeding edge IoT devices and platforms, there’s a strong demand for that,” Mercer said. “But data encryption becomes a priority on those rugged endpoints. People are coming to prefer download-when-available configurations. But at the edge, integrity to these warfighters is the biggest concern. And data tampering is in the realm of possible. A high level of encryption and along with zero trust would need to be implemented.”

Putting the data to work

When the data is available to everyone, and it’s secure, that’s when it can start to be put to use. But the sheer volume of data at this point requires artificial intelligence and machine learning capabilities to be able to parse it. Good data-based decision making requires curated, trusted data.

Because this data isn’t even all in the same format. In fact, a diversity of data sources is key to maximizing and enabling AI and machine learning. Streaming data from sensor platforms, complex data from research and development, highly transactional data from defense business systems can all be brought together to give commanders the ability to designate their priorities, even if those change during the course of operations.

“Fundamentally, as data becomes more accessible, there is a deeper level of battlespace awareness, and the ability to bring in a mixture of viewpoints to make a decision,” Mercer said. “That’s always a good thing. If they have more data, and they have more diversity in the data that’s coming to them or at their fingertips, that’s going to make a better informed decision in the digital battlespace, and that’s always a good thing.”

Copyright © 2024 Federal News Network. All rights reserved. This website is not intended for users located within the European Economic Area.