Our vision of a telecommunications SOA is that it will be an open, federated, secure architecture that allows the subscribers to choose the best-of-breed services from competing service providers and have these services work cooperatively.

The telecommunications SOA has to be federated. Services on it will require aggregation across autonomous boundaries. Take, for instance, using the presence service as a prerequisite to establishing a voice or video session between two participants. The presence state of the participants may be distributed across multiple devices (cellular phones, laptop computers, personal digital assistants, desktop computers) and service providers (Yahoo!, AOL, MSN, enterprise networks). Intelligence will be required to aggregate the presence and availability of a subscriber across states maintained in different autonomous systems. We note that grid computing, with its notion of virtual organizations, is well suited to solve portions of this puzzle. Further research is needed on the application of grid computing to telecommunications SOA.

The telecommunications SOA has to be built on open protocols. Not only will this aid in communications between federations, but also it will allow subscribers to describe their preferences and devices to enunciate their capabilities and store all of these in a secure repository accessible by trusted services. Open protocols will also allow devices to negotiate and prioritize services among themselves. In short, a robust discovery, selection, and binding mechanism for services in the telecommunications SOA will be another area of active research.

The telecommunications SOA has to be built with pervasive security. Devices must not be allowed to update a repository unless they can prove their authenticity. A subscriber must not be allowed to set up a session with another unless he can prove his veracity. Authentication and data integrity are that much more important in a telecommunications SOA because of the need for privacy that we accord to our communication needs.

And finally, the telecommunications SOA must support a multiplicity of service providers. In the Internet domain, innovation is played out as various providers — AOL, Google, Yahoo! — attempt to make themselves the preferred portal for a Web surfer by providing attractive services. Google, in particular, has played this card well. A Web surfer can use Google not only for browsing realty listings, but also to contact the listing realtor using Google Talk, take a virtual tour of the area in which the property is located using Google Earth, and then find directions to the property using Google Maps. Further innovation in the telecommunications domain will occur when multiple service providers vie for the attention of a subscriber. Providing extensible SOA frameworks for telecommunication services continues to be another area of fruitful research.

The evolution of a telecommunications SOA has just begun. The rest of this book provides an early glimpse into this evolution.

Chapter 3

0 0

Post a comment