INTEL?PROACTIVE COMPUTING

Prev don't be afraid of buying books Next

INTEL—PROACTIVE COMPUTING

Another project of note is Intel's research into proactive computing. A new era of computing is on the horizon. In this new era, billions of computing devices will be deeply embedded within our physical environment. These tiny sensors and actuators will silently serve us, acquiring and acting on a multitude of data to improve our lives and make us more productive. However, today's model of interactive computing, whereby individuals interact one-on-one with computers, inputting commands and waiting for responses, will not scale to the new environment.

The interactive computing model is already showing its limitations, as we begin to confront the challenges of dealing with multiple computers, from desktop and laptop systems to cell phones, PDAs, and a growing variety of consumer electronic devices. When we each have hundreds or thousands of devices to deal with, it will be impossible for us to interact directly with each one. Intel proposes that the time has come to transition from interactive to proactive computing. Proactive computers will anticipate our needs and take action on our behalf. Instead of serving as glorified input/output devices, humans will be freed to focus on higher-level functions.

Intel has identified three steps that are essential to making proactive computing a reality. The first is getting physical—connecting billions of computing devices directly to the physical world around them so that human beings are no longer their principal I/O devices. The next step is getting real—having computers running in real time or even ahead of real time, anticipating human needs rather than simply responding to them; early examples include airbags and antilock brakes. The third step is getting out—extending the role of computers from the office and home into the world around us and into new application domains.

Small, inexpensive, low-powered sensors and actuators, deeply embedded in our physical environment, can be deployed in large numbers, interacting and forming networks to communicate, adapt, and coordinate higher-level tasks. As we network these micro devices, we'll be pushing the Internet not just into different locations but deep into the embedded platforms within each location. This will enable us to achieve hundredfold increases in the size of the Internet beyond the growth we're already anticipating. And it will require new and different methods of networking devices to one another and to the Internet.

Increased access to data is a necessary, though not sufficient, condition if we are to obtain greater productivity gains from human endeavors. As we move up the stack and integrate data from the world around us, we must learn to deal with uncertainty. The physical world does not exhibit the deterministic behaviors computer scientists have come to know, model, and love.

Anticipation

In the era of proactive computing, software will anticipate our needs. Excess computation and communication capacity will be harnessed to fetch and manipulate information, producing answers before they are required—much as a chess champion predicts his opponent's moves many steps into the future. We see glimpses of anticipation today, both in the speculative execution features of Intel's processors and in some of the most advanced Web proxy engines. Intel's project is related to the emergence of a new genre of machine learning tools that is firmly grounded in statistical methods. Systems such as those under development by Stanford, University of Washington, and Carnegie Mellon University exploit uncertainty to support new software techniques, a key stepping stone to anticipation.

Once proactive computing has learned to manage uncertainty and can get computers to anticipate needs, the next challenge is bridging the gap between anticipating those needs and acting on them. The difficult step is not so much in determining the action to be taken, but to develop the feedback and control mechanisms essential to the stable operation of any closed loop system. Today, most systems rely on human beings to close the loop and provide stability, placing human beings under intolerable stress.

Figure 14.3 summarizes the research challenges facing this project.

Figure 14.3. A summary of the Intel Proactive Project.

graphics/14fig03.gif




Amazon


Autonomic Computing
Autonomic Computing
ISBN: 013144025X
EAN: 2147483647
Year: 2004
Pages: 254
Authors: Richard Murch

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net