Over the past five years, there has been an unprecedented advancement in computing, especially in hardware size and speed. For example, in high-speed computing, a cost/performance goal of 200 billion operations per second per million dollars has been reached. This represents a tenfold improvement compared to just 10 years ago. However, by 2002, this value is expected to reach 2.5 trillion operations/second per million dollars and by 2010 the figure could reach a staggering 360 trillion operations/second per million dollars.
This represents an incredible advancement in high-speed computing. However, the current size of this computer is eight standard racks of equipment. This is still too large for military applications, especially in a tactical environment.
Numerous efforts are ongoing to further miniaturize hardware systems. For example, great strides have been made in the miniaturization of memory devices. By 2002, it is expected that the technology will be available to shrink memory devices down to the DNA level.
This means that one will be able to store 1018 bits of information in a system occupying only 1 cubic centimeter of space. This is equivalent to placing all the libraries in the United States, including the Library of Congress, in one single cube.
These advances in memory storage will allow the further shrinkage of large high-speed computer systems. By 2006, it is estimated that the high-speed computer that is currently housed in eight racks will shrink down to the size of a kitchen toaster.
The research pushing the reduction in size and the increase in speed of computing systems has been critical to the U.S. Air Force's ongoing "migration to space" as an investment area. However, even all these advances are not nearly sufficient to meet the Air Force's advanced-computing needs for 21st century space-based operations.
For example, the Air Force Space Command has five major mission areas: space control, Force enhancement, Force applications, space support and mission support. In support of these major mission areas, the Air Force Space Command has designated 183 technology needs, of which 71 (or 38.8 percent) have associated computing/size implications. The requirement translated into technology specifications represents the next generation of size/speed requirements.
Given advancements in space systems such as high-data-rate hyperspectral and ultraspectral sensors, by 2010 the computing requirements will be greater than the estimated 360 trillion operations per second per million dollars forecast.
To put this all together into a single, integrated concept, the Air Force Research Laboratory's Information Directorate has embarked on a new initiative called the Joint Battlespace Infosphere (JBI). The JBI is a developmental paradigm that will form a seamless tie between data converted to knowledge and users. The integration of advanced-computing requirements with the JBI will be the key to an integrated, predominantly space-based, operational command-and-control system for the U.S. Air Force of the 21st century.
The Joint Battlespace Infosphere can be defined as a globally interoperable information "space" that aggregates, integrates, fuses and intelligently disseminates all relevant battlespace knowledge to support effective decision making at all echelons of a joint task force. It will constitute a seamlessly accessible combat information management infrastructure linking all sensors, systems and users in a joint task force to achieve informational unity.
Not intended to replace existing command-and-control systems, it serves as the integrating substrate upon which existing and emerging systems will be linked together to support transparent information exchange across a full spectrum of mission activities.
One can think of the JBI more as a philosophy of doing business rather than a single weapon system. With the increased computing requirements that must be taken into account during the development of the JBI, the Air Force Research Laboratory is embarking on a "migration to space" concept. This concept will eventually shift science and technology dollars from predominantly airborne systems to airborne-and-space science and technology areas. One challenging area is in the development of smaller, cheaper satellite systems.
Small as a softball
These new space systems (called microsats or nanosats) could be as small as a softball, yet perform functions that are currently being done by large, vulnerable ground-based systems. It is envisioned that these nanosats will fly in clusters, with each separate satellite performing a subset of the total mission. This establishes two of the most challenging computing requirements: formation flying and distributed control.
Imagine 50 nanosats flying in formation at 17,000 mph while their spatial orientation has to be measured in inches.
The advancements in computing power, coupled with ever decreasing sizes, have allowed further advancements in workstation computing power. Advanced sensor capabilities (for example, hyperspectral and ultraspectral), information fusion and the ability to move the mission ground station to space are but three examples of where advancements in computing will play a critical role in advancing an expanded JBI concept.
New strides will require numerous new technology paths that must be followed. Some examples include intelligent agents for autonomous operations; knowledge-based information systems for smart data retrieval; information visualization of real-time data; embedded architectures for command and control on a portable device; and phenomenological models for command and control to allow decision makers to "smell" the battlefield.