Build Web Services from existing applications
What will happen when a new idea be a really good thing? People compete for claiming that they made this idea. And another group of people want to occupy this idea and add or modify what to do so, so they can say that they make it better.
Maybe this is the situation of Web Services. IBM, Microsoft, Sun, HP, Oracle and BEA are participants, they or claims that they have created Web Services or they can provide the best Web Services services.
Why do all these sellers want to lead this field? The answer is: money. Web Services is a great idea that allows applications to be integrated together and deploy based on applications, which are built on blocks, blocks, or API follow self-certification standards. This concept will infer the concept of how people develop and deliver software applications.
According to Gartner said, the Web Services market will expect to grow to 28 billion in next three years, but Web Services is not just a good idea of the trend, nor is it about how software will be developed. It will also have how software is sold, and more important is how we will purchase and pay for it. In fact, it is reasonable to think that Microsoft is moving the entire software technology to a main reason for the .NET-based Web Services model is a chance to give the company to fully change its business model for the application.
Defining Web Services Behind Web Services is that software solutions can be delivered to discrete services on remote Web, using universal ways to understand each service, what is integrated with it. In order to achieve this, the industry has established some standards such as unified description, discovery, and integration (UDDI), Web Services Definition Language (WSDL), and Simple Object Access Protocol (SOAP). These criteria let an app developer find a web service and understand how this web service is running and integrates this Web Service, which effectively uses remote calls based on standard interface language.
However, the functional module (probably anonymous) of the host (may be anonymous) is still far from the reality. There are many problems to overcome before these fantasies become reality. Here are three main obstacles:
• There is no SERVICES - this may be a bit exaggerated, but it accurately applies services for our exquisite pre-built-in-factor. The industry is created many tools we can use to develop new Web Services, and IBM's Web Services tool is to help us develop a positive example of new Web Services.
• There is still no business model - no commercial model that doesn't really make it. Just like ASP concept three years ago, web services is a good idea, but how do you use it to make money?
• The shared motivation may not sufficient - the drive power after Web Services is to share their intellectual property in order. Unfortunately, the internal function of the software is seen as a private service. Therefore, it is aware that they sell some internal features in the market rather than a complete solution, software development companies will vome to transform their ideas.
Why is there a panic? Then, if Web Services is still in the young stage, why can we hear so much about it? Because it is advantageous! The idea of Web Services is now in use, but is not as used as the source of application functions, but integrates the application together and can build new applications with high-yield upgrades between enterprises or trade partner organizations.
Two Factors Driven Development Enterprise Application (EAI): • Scalability / Performance
• Business flow granularity control within integrated system
EAI's development has turned from batch update procedures to integrates the application level of things, and then uses high-level messages to the application integration. The next demand will be to achieve greater scalability and the granular control of the integrated application of the business function level, while the Web Services architecture provides a method.
Today, most of the Web Services architecture project is an EAI project that builds application services within an enterprise or trade partner. The company is realizing that in order to develop applications, if they want to move to the Web Services architecture, they need a way to convert to new Web Services in the enterprise. This choice (will be rewritten in some gold function module as a new web service) is low-cost.
Imagine such a scenario: a business offers all of the viable logic of its core commercial application as a reusable web service. In other ways, as a need for new application software development, developers can immediately call the rich web services to become new applications. They can focus on striving to develop new business process services, which can then be developed in the future.
Many companies recognize that the assembly of legacy applying business guidelines is a very valuable investment. Web Services provides architectures and a set of standards that become these project structures.
Function point analysis In order to measure the investment in productivity and software development, many organizations use functional point analysis. The advantage of using functional points is to focus on metrics and their programming language and independent technologies. The average of each programmer has developed 270 new application function points for one year.
Most of the core applications consist of tens of thousands of functionals. Developing a new application function to spend $ 550 to $ 2,000, or even more, and the core system of existence may have millions of food investments, so a forced choice is to reuse methodology, let you develop investment Continue to harvest returns and reduce the risk of spending and new project development.
Publish your core application logic as reusable Web Services to get investment income, in order to transform your company into flexible, high-integrated technology. In addition, this reuse process will help remove knowledge islands in your IT company and help you optimize new project resources. Your legacy application development resource will be absorbed by Java or .NET programmers.
Three steps to successfully include three steps for Web Services for Web Services:
• Know what you have - many strategic applications have been developed for decades, modified, so documentation is outdated or does not exist. Often, this is the most important application has become unknown field. If you rash, there will be a lot of trouble. You may display a message: "If it doesn't stop, don't repair it." Only "Ancient noble" COBOL programmer can involve this area. Unfortunately, when those cobol programmers leave or retire, many companies can only look at COBOL programming skills, and it is difficult to replace these individualists and knowledge.
So, if we want to turn your legacy to dynamic, easy to use and easy to understand Web Services, what we must learn to legacy? What do they do? Where is the business guidelines in them? You can use the app understanding tool to do this. Do not misunderstand your application understanding tools as code restrictions for Y2K correction projects. This tool you need must provide a whole and semantic understanding of your application. When an internal application stream, event processing, and business logic are provided, it must understand the context of the entire application, including program code, database, and work control environment.
• Dudout Real Gold - Once you understand your core system know where the business criterion is, then the next step is to isolate them for reuse of Web Services. This process can be divided into non-intervention and interventional ways. Sometimes it is difficult to separate code elements associated with a special business function. Less interventions may make more sense, this method will publish the business process of Web Services to functionality, which can be accessed by screening or from the database creates a service. Simple screen erasing may not provide a Service-based architecture that requires the publication of actual business features to Web Services. Any non-intervention based on the screen I / O interface must provide this level of granularity to publish that each screen interface is an application Services collection, or any screen group is a single-use service. The interface-based screen provides optimized scalability is also basic. If each process is based on the screen service, the architecture is based on the relationship between peers, then it is impossible to be a full solution.
Establishing a database Services may also be a challenge because many commercial systems are based on a COBOL file system or other non-relational data. Access to pure data based on architecture-based services means that these data is exposed by some form accessing languages (such as SQL). Most COBOL databases do not support local SQL access. However, there is an intermediate product in the market to provide open access to non-SQL adaptation data, then these middleware tools need to be integrated into the Web Services structure and standard.
More complete and most strategic solutions are publishing program logic for reusable services, which requires an intervention method - enter the code. If you invest enough investment in applying understanding, then this is not so difficult. Moreover, the available tools can help you isolate and extract data from the existence program, then package the data as a discrete Service.
• Create Web Services - After you understand your application, then discover and gain you want to reuse business guidelines for Web Services, you need to adjust your products so you can publish Cobol, C or other legacy program for SOAP WEB SERVICES.
An optional adaptation method is to automatically translate separate data into Java or C #, aware of this, although it sounds attractive, but automatically converts legacy language code to adaptive web services code, it has a lot Defective. You can hardly perform clearly in new languages. Typically, the original code result is an incorrect architecture in a new target language. You get its semi-formed results, no matter what, your programmer should understand or continue to work.
If the adaptation method stays in the existing language, a good adaptation environment will allow you to use the new language written in Services and the remaining Services mix. This evolutionary method makes you gradually conversion of fundamental technologies, avoiding the risk of reproduction of disastrous disasters.
For the right implementation of these three steps, you will need software tools and middleware (see Figure 1) to help you get your work.
Figure 1 is a midnovation stack for transforming applications commercial guidelines for Web Services
Conclusion The above process describes an effective way to create web services from existing core business applications. For its benefits, only brave companies can start with a big project to translate them to the Web Services warehouse in existence. Business needs and drivers can lead these projects.
