UDDI - Universal Description, Discovery and Integration

The protocol and the technical description of a Web Service do not suffice for a widespread embracement of this technology. Programmers know how to access services but a business itself still needs a way to discover other Web Services and expose its own ones. The discoverability of an offered Web Service is one of the main keys to its success.

This is where UDDI comes into place. UDDI is an abbreviation for ``Universal Description, Discovery and Integration''. It is the first cross-industry attempt to address the current limitations afflicting the spread of Web Services. UDDI is not a W3C standard²² but uses many technologies standardized by W3C like SOAP, XML and XML-Schema and other open standards, like HTTP. UDDI is not limited to certain systems, instead it embraces diverse platforms, operating systems and languages that exist on the Internet.

UDDI does not have an important role in this thesis as the discover-ability of fieldbus Web Services is not very important because programmers who access these services already know where to find them. Nevertheless some basic knowledge of UDDI is needed to gain a better understanding of the meaning and the sense of Web Services themselves.

UDDI can hold a lot of information about a business. Details about a specific Web Service are just one part although more information is needed to find this specific service. All information in UDDI is stored in the so-called ``UDDI Business Registry'' as shown in figure 14. As this information is so diverse, it is broken down into three logical parts:

White Pages:

The white pages contain basic business information, like the business name, its address, telephone number etc.

Yellow Pages:

These pages contain further information about the business.

Green Pages:

All technical information goes into this part. This information includes the accessibility of a Web Service, like URLs, method names, the location of WSDL files etc.

Figure 14: The UDDI Business Registry

UDDI Business Registry is stored on so-called operator sites. To improve reliability, each operator site replicates their data to the other operator sites.

Access to the UDDI Business Registry is separated in two basic query types:

1. Inquiry:

These functions can be used for searching or browsing the registry after specific data like businesses or Web Services.

2. Publish:

These functions allow administrators to create, delete or change registry entries.

UDDI also defines error handling, which is done by SOAP Fault messages.

More information about UDDI can be found in [WAL02].