Navy takes steps to boost operational IT
Navy looks for efficiencies in business IT
Terry Halvorsen has been the Navy Department's CIO for slightly less than a year. A member of the Senior Executive Service since 2004, he was most recently deputy commander of Navy Cyber Forces and deputy commander of the Naval Network Warfare Command.
He spoke to Defense Systems Editor-in-Chief Barry Rosenberg about cuts to IT costs — without affecting operations — Next Generation Enterprise Network progress and enterprise e-mail.
DS: What’s at the top of your to-do list?
Halvorsen: IT efficiencies. That is the No. 1 thing on my agenda. Here’s what I can tell you. I don’t think it’s any secret that the Defense Department is going to take budget cuts. So we have a couple choices, and you can jump immediately saying, “OK, we’re going to cut this capability,” or you can look at new ways of doing things that could potentially save you money and maybe even improve your performance. I think IT is one of those areas where we have that possibility, where we could actually save some money and, in the end, maybe even improve performance with very little risk.
Navy works to get IT systems shipshape
DS: You’ve talked about cutting 25 percent out of the enterprise IT budget, focusing on cutting business IT systems and not operational IT systems. Can you attain those IT efficiencies in that manner?
Halvorsen: I’m fully aware that inside of DOD there is not a wall between what is business IT and what is operational IT. But there are some systems that are predominantly about business functions.
DS: You mean like financials?
Halvorsen: Yes. You can look at certain systems and say, “This system is predominantly delivering business data; can we apply some of the business lessons learned in IT to that?” Sometimes the answer will be no; sometimes the answer is yes. When yes, we need to take advantage of that.
DS: So you think there are more opportunities to find cuts in business IT than in operational IT?
Halvorsen: I wouldn’t say there are more. I would say it’s a safer set of opportunities to look at when you can try to stay away from the tactical and operational systems. Take the risk level, for example. If I make a mistake on something that is predominantly a business system, then I have a business problem. Well, that’s bad, but it’s not catastrophic because the price I pay on a business system is generally money and time. A mistake made on an operational system could be about people.
DS: Yes, but inefficient business IT costs money that could have gone to improve the operational side of things.
Halvorsen: Well, Barry, you just caught the other thing that makes this a good option. Any money that I can save on the business side does two things for me: It either directly increases what I can spend on the operational side, or if we do have less funding, it’s where I can take that money from so that I can keep the operational funding where it needs to be to ensure we have a good force.
DS: So what are the operational funding priorities?
Halvorsen: There are a couple things that we need to do. One, whether it’s bandwidth or new technology that enables us to pass more data over the same bandwidth, we need to make sure that the data that’s needed at the edge — be that an at-sea edge or a shore-based edge — can get where it needs to go in a timely way.
And it’s going to take not just technology but some detailed analysis to say what really is the data that needs to be at that tactical edge and on what timeline does it need to be there. That may be the harder problem. How do you dice your data? How do you set some policies and standards that say: "This data meets the criteria and does need that millisecond of time to be delivered to a forward edge command"?
DS: Moving on to NGEN, the continuity-of-services contract extends the time at which NGEN is to be implemented until 2015. How is the Navy CIO office taking advantage of this additional time?
Halvorsen: The continuity-of-service contract was a good bridge to give us the time to talk about what NGEN is going to be. So here’s what I think NGEN is going to be. I’m going to take a little risk and say that NGEN is going to deliver today’s requirements knowing that, in the IT business, the requirements will change somewhat but the basic functions won’t.
I need connectivity. I need e-mail. I need the ability for people to collaborate. People are going to need to get to something like Microsoft Office or a business suite of tools. We’re going to have that, but we’re going to have and take advantage of [what] I’m going to call state-of-the-shelf technology.
DS: State of the shelf?
Halvorsen: State of the shelf. We don’t want to be, I don’t think, particularly in networks like NGEN and some of the more business-centric systems, on the leading edge there. NGEN does more than business, as it is somewhat a C2 system, but it’s generally about C2 of business functions.
We’re a big business enterprise, and that’s a dangerous place for us to be in terms of cost. I want to be using the latest proven technologies. That’s where I need to be. That’s kind of that sweet spot that as a big enterprise we need to be in. Sometimes it’s hard to find. So that’s what we’re going to try to drive NGEN to be.
It’s going to mean that we take a look at things like — and I hate the word “cloud computing”, but I will use it today — how can we take advantage of the cloud.
You saw the announcement that the [General Services Administration] is the first government agency to have their 17,000 customers and e-mail now in the cloud. I don’t know that we can because of our bulk go to full cloud, but we want to entertain that thought. Maybe we can do a cloud-based computing system for e-mail. It certainly would lower some of the infrastructure and cost.
GSA is quoting about 17,000 customers, and we have obviously a lot more than 17,000. We have about 800,000 customers between the Marine Corps and the Navy. So GSA, by using a commercial cloud-based e-mail system, has said that they’ve improved access and that people can access their e-mail more readily, which I think they’re saying means people can access it from their homes or wherever they are.
So one of the other things we’ve got to do is to figure out how we do this. They say that they’re going to save $15.2 million over the next five years for those 17,000 people. There’s a big savings if we can work out some of the additional security requirements we have on the unclassified side. None of this I’m talking about applies to the classified piece of our business. That has different set of things we’ve got to worry about.
NGEN is going to deliver in . That’s the first time we really get NGEN. The continuity-of-services contract runs through '14. We put the RFIs out. So what I’m trying to do here is say, “In '15, this will probably be state of the shelf.” It is state of the shelf in some areas now.
DS: Let’s move to a related subject: enterprise e-mail, specifically the Army’s migration of its e-mail to the Defense Information Systems Agency. What are your views about the direction you want the Navy to go as far as participating in the larger enterprise architecture?
Halvorsen: I will just say this. I think in the end game, we ought to be able to have a DOD-wide architecture that would support a common e-mail distribution system and would certainly support a system that may let us share common applications. It’s no secret that all of us use Microsoft Office today. I don’t know if it’ll stay there forever, but we use it today.
We’re going to have to have something like that to do our basic functions such as business spreadsheets. At some point, we ought to have a common DOD structure for that. We’re not there yet, though. Right now the architectures of the services are different. I don’t think it’s possible to go from what we have today to an immediate jump to a common architecture.
And I define common architecture at a level of technical spec that lets it work. We can say we’re going to share data, and if that’s your definition of common architecture, then we got it. We can meet that today.
I think they’re implying a level of architecture that’s beyond just sharing data. It really does mean some common applications, maybe a common e-mail system. So to get there, it’s my belief what we’re going to have to do is take that in some stages.
DS: Isn’t Army e-mail migration to DISA a stage?
Halvorsen: I think that’s a stage. But I’ll say this: the Department of Navy today has an enterprise network that does common e-mail across two services and serves 800,000 customers. So I think I’m on that right target. Nobody else has that yet. So as the other services get there, and as we look ahead to see what we can do, I think the first thing we can get to would be a federated set of e-mail that allow us at the user level to operate fairly seamlessly, though they’ll be a lot of back-end processes going on. And then the follow-up step would be maybe some type of common technical architecture that would let us get to a single e-mail system.
One of the fears I have about that, and it’s one that we’ve learned in managing NMCI, is that size absolutely has an importance here. With NMCI, one of the things that we found — and we broke ground on this — was that nobody had modeled anything that size. So when people say that we can distribute changes to the software to a whole network in a day, you know what, even when using the best stuff, with 800,000 users you don’t get there in a day.
You really have to think about what does that size do for you. Is there a new technology or maybe a new way of thinking about what it means to be a federated network that gets you all of the capabilities you want, but still segments the network enough that you can quickly do maintenance and other things you need to do to the network?