Abstract

In order to develop appropriate solutions for embedded device networking, we must understand the benefits offered to the end user of the device as well as the costs involved with delivering a solution. As proponents of networking technology, it is tempting to overestimate the perceived value of connectivity, while at the same time overlook the hidden costs in implementation. It is important to remember that companies adopting technology for embedded device networking are manufacturers who tend to be very conservative due to narrow margins and fierce competition.

A prerequisite to wide adoption of this networking technology is a convincing case for a strong return on investment. Therefor, all optimism must temporarily be put aside for a critical cost/benefits analysis that will provide criteria for judging the suitability of proposed solutions.

Since there are so many embedded applications, it is not surprising that quite a few applications are 'no-brainers' which require little deliberation as to value or implementation method. Administration of networking hardware, for example, is increasingly accomplished through Internet-based device interfaces.

Internet standards and associated technologies provide a remarkable set of opportunities for enhancing the value of existing and new embedded products. These Internet technologies have been available to millions of large, 32- and 64-bit systems for some time. The key challenge we have undertaken is to establish a distributed device-networking platform that provides appropriate solutions for all embedded devices. Our goal is to make connectivity practical even for systems that do not have convenient networking already available and are severely constrained by economic pressures.

• View the full text of this paper in HTML form and PDF form.

• If you need the latest Adobe Acrobat Reader, you can download it from Adobe's site.

• To become a USENIX Member, please see our Membership Information.