 |
|
|
 |
||
 |
||
 |
||
 |
||
 |
||
 |
 |
|
 |
||
 |
||
 |
||
 |
||
 |
||
|
||
 |
Your independent resource on business integration and network computing through middleware and message brokering
Network computing and middleware for the next generation
Steve Ross-Talbot
Chief Technology Officer
SpiritSoft
Management introduction
The ability to communicate, to interact, is one of the principle differentiators between man and the rest of the animal kingdom. The successive technologies of the telegraph and telephone heralded unprecedented growth in the world economy. With the arrival of computer systems and integration of telecommunications - wired and wireless - the capability to communicate meaning between people, enterprises and systems has never been more important.
Simultaneously, the growth in the complexity of IT infrastructure has given rise to unprecedented growth in integration software. From databases in the 1980s to client/server in the early 1990s to the Web in the mid 1990s and now Web Services, integration tools play a major role. As application server vendors and messaging vendors start to consolidate, so network vendors are looking to add more value. What do we have to look forward to? Predictions are dangerous. Divining the future is certainly less sure than explaining the past. Nevertheless, looking at the foci in academia, in JCP, in the W3C and in Web Services, certain patterns emerge.
In this analysis, Steve Ross-Talbot looks at the implications of the autonomic computing initiatives, rules, intelligent networking and Web Services. He shows that they are all interrelated parts of middleware. Autonomic computing may be a long way off; nevertheless, all the other technologies are playing their part in furthering the aims of autonomic computing - and some of them are available today.
Figure 4.1: RuleML as an open standard
Figure 4.2: The RuleML interchange model
Figure 4.3: Useful references
Management conclusion
Autonomic computing has, rightly, been used by IBM and others to illustrate what future computing needs. In comparing it to what exists a number of technologies have been highlighted, each of which has a role to play in delivering autonomic solutions.
In this context, JAIN and PARLAY are moving in a direction that will ensure the networks are transparent. Mr Ross- Talbot also describes research in the use of soft switches that can be configured while they are running by using rule technology (RuleML). This, as he infers, is as a means of encodifying reactive and deductive behavior. Its influence, and the relevance of Web Services running on intelligent networks, has also been identified.
Whilst wholly autonomic solutions are some way into the future, the technologies inherent in JAIN and PARLAY provide a solid bedrock for network independence. RuleML then provides a clear way forward for the adoption of rules for soft switches as well as for orchestrating business logic and providing adaptive management of applications, many of which will be built around Web Services.
All in all, for those who thought that middleware was reaching a mid-life crisis, as the enthusiasm for EAI matures, the truth can be seen as very different. Middleware is growing up, flexing its wings and being extended into new process realms.