Information Dominance Bows to Network Limitations
The warfighters’ focus must be on the mission and not on the performance of the infrastructure.
Incompatibility of applications compounds the problem. More than seven million information technology devices running more than 7,000 major applications reside on the defense networks, but it is rare for systems in finance, personnel or logistics to share even the most basic data elements. In addition, the defense networks do not possess assured security enforcement or enterprisewide configuration visibility.
For the warfighter to achieve information dominance, the Defense Department must transform its loose set of independent networks into an interoperable, a secure and a cost-effective system for collecting, processing, fusing and displaying text, graphics, video and scientific data.
The newly integrated Office of the Deputy Chief of Naval Operations for Information Dominance (N2/N6) has defined information dominance as having four key principles: Every platform is a sensor; every sensor is networked; every collector and sensor will be dynamically tasked and managed; and every shooter must be capable of using target data derived from any sensor.
These principles have an important difference compared to other information-centric principles and frameworks that have been offered over the last two decades. This definition positions information dominance technologies as weapons that need to be integrated with the other weapons in the military arsenal. This acknowledges that cyberspace represents a military theater with its own sensors, platforms and warfighters. Computer technology and computer networks become merely the means to transport information. It is only when sensors are fused with information flows that personal intelligence can be applied to making decisions.
But these principles also reassign information technology and information networks into an infrastructure role. They force the Defense Department to revise its thinking about how it conducts its operations. No longer is it possible to depend solely on network-centric connectivity based on the Global Information Grid (GIG). Having a dial tone is insufficient. Instead, we need a framework that delivers sensor-to-shooter useful information in real time. The need is for networks that are so predictable and so reliable that nobody needs to worry about the technical performance of the department’s infrastructure. Instead, the military can concentrate on what is relevant for combat.
Today’s defense information technology infrastructure consumes more than 50 percent of the total department budget and employs more than 300,000 support personnel. That is too labor intensive and highly error prone. Simultaneously decreasing costs while achieving Google-like ease-of-use and 99.999 percent uptime, which equates to roughly 5 minutes of downtime per year, will require change. Systems management responsibility must shift away from information technologists, who largely are contractors, to warfighters, who have to make the trade-offs between costs and network services. Today’s stove-piped solutions must be eliminated, and this can occur by imposing enterprise standards. The department must rethink its approach to information technology governance so that user requirements are met, rather than today’s approach of forcing users to accept what the information technologists can afford to deliver.
None of this thinking is new. There has been no shortage of ideas for more than 20 years. The problem is that despite an ample supply of concepts, the department still is not achieving information dominance.
The visions of past department leaders have not as yet materialized in the ways in which defense is delivering information dominance. For further progress, the following issues must be resolved.
The majority of the existing 15,000 defense networks depend on the Internet. Even the GIG relies on services delivered by contractors who depend on links, routers and switches that are vulnerable to attacks. Though elaborate security precautions have been taken for security assurance, it is questionable whether the military can completely send critical traffic from sensors to shooters over the Internet without a compromise.
All critical communications as well as all messages that can reveal military intent must be conveyed over Internet networks, routers and switches that are owned and operated by the military. Companies such as Google, Akamai and banking firms already have such capabilities at costs that are affordable.
Network costs are greater than information technology costs. Over the last 20 years, total defense network costs have more than doubled while the cost of information operations was cut by more than half. The costs of computer capital have been reduced by 80 percent for identical performance. At the same time defense manpower has been declining.
The total defense spending that is available for networks continues to decline because labor costs are escalating. For this reason the funding for information dominance will have to come from personnel cost reductions. Except for a few recent instances of virtualization of data centers, the management of defense processes does not favor manpower cuts. For cost reductions, it will be necessary to automate the management of seven million desktops as well as to monitor for improved security through cloud computing, which offers quick paybacks. Contractors can operate such networks, but the network control centers and data centers must remain under the cognizance of the Defense Department.
Business operations that account for almost a third of total costs also should be benchmarked against commercial firms. The information management costs of the Defense Department are more than 10 times greater than comparable expenses by the largest commercial firms. The cost of the enormous defense infrastructure should benefit from the economies of scale.
Based on Navy ratios, for every dollar spent on computers and communications—the information transport—more than 30 cents is spent on sensors. Seeking real-time as well as 99.999 percent reliability of the links between shooters and sensors will change the ways information dominance trade-offs are managed. The roles of the chief information officer will have to be subordinated to the warfighter command structure that starts with U.S. Strategic Command and flows through U.S. Cyber Command to military component cyber commands. Funding then should be allocated on the basis of short-term changes in combat capabilities to support specific missions.
Information dominance is characterized by very short time intervals for making changes. The typical information technology program has a development-to-fielding cycle of many years. An information superiority system must be able to accept major changes in a matter of days. Consequently, the acquisition processes cannot follow the patterns of procurement for weapons that have a very long life but must suit programs that have a life expectancy of a week.
The funding and milestone control of information dominance programs will have to shift away from compliance with directives that placed the acquisition organizations in the lead. The current organizational walls between systems planning, systems development, systems acquisition and systems operation are not viable. The existing systems life cycle process imposes a rigidity of completely separate silos that prevent offering information services that can adapt rapidly. Instead, the defense acquisition system that has been designed for airplanes and ships will have to develop new processes adapted to information management as required by Section 804 of the fiscal year 2010 Defense Authorization Act.
There is no way of delivering information dominance with the current proliferation of tens of thousands of aged computer programs that house thousands of incompatible databases. Military services do not have the funds to support migrations to new standard applications. As an example, the Defense Integrated Military Human Resources System failed to obtain departmentwide acceptance from the services even after 10 years of trying and untold hundreds of millions of dollars. Agreement on an over-arching standard system was not feasible.
For rapid transition to information dominance, the department must follow commercial examples, such as in the case of VISA credit cards that connect, in real time, hundreds of thousands of banks—each with different systems—with millions of credit card readers. VISA sets tight standards for centrally managed data and for central security authorizations so that translation tables can be constructed for instant interoperability between incompatible systems. The department has experience with the use of such mediation programs, for instance the Global Exchange Service (GEX). Real-time GEX can work provided bandwidth is available.
New networks must offer global mobile access to computing and to voice communications by means of portable devices such as smart phones. There should be no distinction or logical separation between messages received at the desktop, desk phone and any portable appliance, regardless of where an individual may be located. A unified approach to secure mobile communications will require the department to install and to manage organic wireless connectivity as a shared enterprise wide offering.
The Navy has forged ahead by advancing information dominance as a concept to guide its operations in the era of cyberwarfare. It has combined $5.6 billion of computer and network spending with the costs of $3.9 billion of sensors. The corresponding manpower costs are greater than $5 billion for a total budget in excess of $100 billion.
This combination of intelligence, information technology and sensors offers a totally different way to manage information in the future. It is noteworthy that the chief of naval operations has approved the Information Dominance Corps Warfare insignia. This means that information warfare is now in the same categories as surface warfare, aviators and submarines.
The defense networks of the future must be shooter-centric. They must support not only lethal combat but also associated tactical maneuvers. Such systems must be not only reactive but also anticipatory of the adversaries’ countermeasures. Network-centric systems are insufficient in assisting the military in the extraction of data from thousands of databases for delivery of improvised displays that are usable in real time.
Every Defense Department shooter becomes the juncture for which networks, databases and applications must operate securely and reliably. Networks are designed only after the connections between the shooter and the relevant intelligence are established. The design of network starts with the enablement of people, such as shooters, rather than with the technical means for transporting information—for example, network-centric delivery.
Instead of a network-centric GIG, the time has come to augment this concept with the capacity to deliver universal interoperability connections that can adapt to future scenarios of warfare that may challenge the department at least 10 years ahead. To accomplish that challenge the department must move on from network-centricity to shooter-centricity.
Paul A. Strassmann is Distinguished Professor at the