Similarly to other companies, BEA also went through the different stages of its evolutionary technological development before it found the solutions deemed to be the ultimate state of the art today. The start was represented by the implementation of J2EE technology, then BEA concentrated on the different standards, while nowadays it already provides an ever improving all-round service for SOA (Service Oriented Architecture) based developments, which is considered to be the general standard in the industry, that is to say - instead of monolit applications - it focuses on application design based on unique services and their publication. At the same time it also means that applications must be designed and implemented in a different manner, and of course different tools are needed for it. Another important aspect of innovation is that these Java tools should operate in a reliable way - explains Iván Péter Ertner, leading expert of Alerant Informatikai Rt., who is in charge of the sales and implementation of BEA products. The significant element of the entire process is reliability. Such application servers and platforms must be developed, which are capable of their own supervision and are automatically restarted in the event of a failure. Services are also distributed to several different locations, so if one of the systems breaks down, the same service functions installed in another system can take over the operation - thus the emphasis is on operability and scalability. These so-called liquid computing computer network concept-based environments are able to respond to even very quickly changing business requirements and they successfully handle issues like bulk administration of server farms or parallel run - that is when several administrator supervises the same server at different points of time -, the manner of auditing and accelerating the speed.
The central element of BEA liquid computing concept - in addition to WebLogic servers - is QuickSilver, which is responsible for the management of corporate services, and which at the same time is the spirit of the next generation of Enterprise Service Bus. Now, what does this mean?
So far it has been difficult to assess and maintain services within the different companies, or to define their operation location, furthermore it has been difficult to find out how to call them, what they implement, etc. This issue has become especially problematic when the existing services were further extended with additional services being dependant on the existing ones, moreover they might have been delivered by different suppliers in case of large organizations, like banks or telecommunication companies. It was also possible to have certain elements within the architecture, which had been provided to the organization readily-developed - for instance an application running on Windows server in Java environment - and these "fell outside the scope" of the standard environment.
QuickSilver, as the focal element of the company's infrastructure, supervises all the services, which are called in via QuickSilver, since it can identify where these services are located within the - maybe extremely complicated - IT environment of the company, and also knows which parameters must be used for calling them in.
Thus the former point-to-point type integration is replaced by the convergent application integration. Its main idea is that the applications are connected to a central service node, this being in charge of trafficking requests to the appropriate service provider and performing the necessary message conversions. The individual applications don't have to take care of giving the parameters in a form the relevant service requires it.
With the help of QuickSilver the co-operation of other applications provided by different vendors can be also resolved without any difficulties. The application using the published services do not need to know on which platform the services are run or who and by using which technology developed them.

Orderly development
At the same time the changes in services do not have any impact on the applications using them; the company is able to more quickly respond to the individual changes, while applications remain stable. This arrangement facilitates a much more efficient implementation, one-point management, monitoring, auditing and development of service based infrastructures.
Such a service-based company can be created, where all the conditions for an orderly development, distribution and aggregation of business logic are in place. All the services and finalized business logic are available in one place within the central infrastructure, this way overlaps between services and the development of separate elements can be eliminated, plus it becomes very simple to develop new platforms and launch higher quality services. The goal is always to significantly reduce the complexity of applications as well as their development cost and time.

Attila Mártonffy