Trufun Kant model-driven architecture

Trufun Kant model-driven architecture

Model Driven Architecture (MDA) is a software development framework defined by OMG. It is a framework based on UML and other industrial standards that supports visualization, storage, and exchange of software design and models. Compared with UML, MDA can create machine-readable and highly abstract models that are stored in a standardized manner independently of the implementation technology. MDA uses modeling language as a programming language rather than a design language. The key to MDA is that the model plays a very important role in software development. Trufun Kant is a very convenient MDA tool developed based on Trufun Plato UML2.x modeling tool, therefore, it can achieve the entire project analysis, design, development process. You can implement a platform-independent model and convert it into a platform-related model. (Note: Trufun KantSeries ProductsTrufun Kant for Java, Trufun Kant for vs.net, Trufun Kant for C ++And covers the features of the above three products and Database ModelingTrufun kant StudioIntegrated Products)
MDA originated from the well-known idea of separating system operation specifications from the details of the way the system uses the underlying platform capabilities. The MDA tool provides a way (through related tools) to standardize an independent platform, standardize the platform, select a specific implementation platform for the system, and convert the system specifications to a specific implementation platform. Trufun Kant is an MDA tool that fully complies with these specifications and provides a unified implementation platform for the system. The three main goals of MDA are: lightweight, interoperability, and reusability through the separation of modularity.
Model-Driven Architecture (MDA) is a new technical system that has been gaining popularity in OMG in recent years. It is also a new hot spot among many researchers engaged in software models, the core idea is to study business models (such as Enterprise Informatization or construction solutions. Then, a domain model with a relative core is extracted and a PIM (platform-independent model) is abstracted ). Then, based on different development platforms (such as. net or Jave), the application platform (windows or unix) forms the corresponding model of the relevant platform (SMS ). The Trufun Kant series of MDA tools have completely generated the corresponding code framework, ensuring the consistency and stability of the implementation framework from the project requirement to the subsequent implementation.
The theory of model-driven architecture (MDA) is still in the exploratory stage. Many tools are gradually improved with the idea of MDA. From the current trend, to fully implement the MDA concept proposed by OMG, it takes at least several years for the MDA tool to become fully mature. Currently, the implementation of the MDA tool is limited to the correspondence between class charts and codes. The code framework is generated, but most tools still cannot implement repeated and reverse synchronization between the two, the Trufun Kant product solves this problem well, so that both the Model Modification and the generated code can be synchronized to the other side in real time, ensure that models and development are consistent throughout the project process.
MDAIt has been applied in the following fields:
* Banking
* Insurance
* Public Enterprises (especially in the Financial Management field)
* Embedded System
* Logistics Support System
Currently, Trufun Kant has been applied in the above fields.
You will also see that MDA does play a role in it. The implementation of MDA is mainly concentrated in the following three steps:
1. First, you use UML to perform highly abstract modeling for your application field. This model has nothing to do with the technology (or underlying technology) that implements it. This model is called platform-independent model (PIM ). Trufun Kant series products are implemented based on Trufun Plato, the latest UML2.1 standard UML modeling tool. Therefore, support for this is beyond doubt.
2. PIM will then be converted into one or more platform-related models (MSPs ). This conversion process is generally automatically implemented. A specific implementation technique will be used to describe your system. The Trufun Kant series automatically converts this process and converts the PIM model to the MSPs related to more than 10 object-oriented languages. This step is the most difficult in the MDA process. It requires you to have a rich and consolidated knowledge of the basic technologies you want to apply. On the other hand, the source model (PIM) sufficient information required to automatically generate the SMS.
3. The source code will be converted to the source code. Trufun Kant products use the bridge technology and code framework technology to implement this function. The code framework can directly convert the model to the code, the bridge ensures Real-Time Synchronization of models and codes.
UseMDAPrerequisites
* A widely accepted fact in the industry (or even the whole world) is that only changes are permanent. Technology is always innovating. This is particularly evident in the middleware field. Of course, database technology, operating systems, and even programming languages are constantly changing. These technologies are significantly faster than the basic concepts in the application field.
* If you work in a specific application field, projects in this field are similar. The entire application family or different projects belong to the same application field. Therefore, the MDA or generation process is especially suitable for you.
MDABenefits
* Your Investment in modeling will be more persistent and effective-far longer than the technology you are currently using to implement it. This will be more conducive to protecting your investment.
* You have technical flexibility.
* You will not be affected by the different change cycles of the technology or application-with the help of MDA, you can maintain the diversity of the two aspects in a neutral manner.
MDADisadvantages
* MDA means more "assembly" rather than "development"-when creating PIM for an application, you basically have no technical cycle space. This is hard to imagine for many developers today.
* The creativity of Software Development degrades to some extent. Developers often think that arguments over a new technology are very attractive at the forefront of technology. However, in the MDA process, a lot of work is to build a model, which is far from the specific technology, but in line with the OMG recommendations.
* Potential immaturity. UML2.x grows at the beginning, and the time for this MDA tool to appear is relatively short, and the future development will continue.
MDASoftware development cycle
In MDA, the software development process is driven by the Modeling Behavior of the software system. The following is the software development cycle of MDA:

There is no major difference between the MDA life cycle and the traditional life cycle. The main difference is that the artifacts created during the development process, including the PIM (Platform Independent Model, Platform-Independent Model, platform-related models) and code. PIM is a model with a high abstraction level and independent implementation technology. PIM is converted to one or more MSPs. PSM is tailored to a specific implementation technology. The final step of development is to change every change to code, which is closely related to the application technology. The transformation from model to model or from model to code in the traditional development process is manually completed. However, in the Trufun Kant series of MDA tools, these transformations are automatically completed by tools. Tools can automatically implement everything from PIM to SMS, and from SMS to code. PIM, SMS, and Code models are used as design artifacts in the software development lifecycle. In traditional development methods, documents and charts are used. What's important is that they represent abstraction of different levels of systems, looking at our systems from different perspectives, and converting high-level PIM to the capabilities of MSPs to improve the abstraction level. This allows developers to have a clearer understanding of the entire system architecture without being "contaminated" by specific implementation technologies. At the same time, it also reduces the workload of developers for complex systems.
In order to improve the efficiency of software development, enhance the portability, collaborative working ability and maintainability of software, as well as the convenience of documentation, the emergence of MDA indicates the way to understand the problem. MDA was predicted by the object-oriented technical community as the most important methodology in the next two years. The main problem with modeling today is that it is only a practical exercise for many enterprises. This causes the model and code to be not synchronized. The Code will be constantly modified, and the model will not be updated, or the requirement will change and the model will be modified, but the code modification is not from the model. In this way, the model and code cannot ensure consistency from the model to the system. If the model is not the final implementation of the Code, the model will lose its meaning, this problem is caused by the development process management problem. On the one hand, some MDA tools cannot ensure the unification of the two, while the Trufun Kant tools have already solved the latter problem, it provides a platform for us to implement this operation. The key to bridging the gap between modeling and development is to turn modeling into an essential part of development and make it a valuable part for us to reuse and reference, it is because our development does not increase the additional cost because of changes in the Code. MDA is a model-driven development framework, and its vision is to define a new way to describe and create a system. MDA makes the purpose of UML go further, not just beautiful pictures. Many experts predict that MDA may lead us to another golden age of software development.
The biggest benefit of MDA is the long-lasting value of the business model, but the cost is to increase the abstraction layer. Trufun provides precise modeling support for the latest version of OCL in terms of modeling technology, more mechanisms can be expanded to support precise modeling and analysis models. You can also customize the business model based on your actual needs.
MDARelated Standards
To realize the grand conception of MDA, OMG has developed a series of standards:
UML: UML is used by MDA to describe various models. It is not born for MDA, but as the most popular modeling language, UML has occupied 90% of the market share in the global modeling language field and has become the de facto standard of modeling language, therefore, it is a natural and wise choice for OMG to use it as the basis of MDA technology. It is the basis of MDA and the most powerful weapon of MDA.
MOF: MOF (Meta Object Facility Meta Object mechanism) is a high-level abstraction than UML. It aims to describe UML extensions or other UML-like modeling languages that may emerge in the future. Although MOF is not born for MDA, we can appreciate the painstaking efforts and long-term eyes of OMG engineers.
XMI: XMI (XML-based metadata Interchange) is an XML-based metadata Exchange. It defines an XML-based data exchange format for various models through standardized XML Document formats and DTDs (Document Type Definitions. This makes it very important to pass the final product model in different tools, it ensures that the MDA will not be added with a new constraint after it breaks a constraint. Trufun Kant allows you to generate standard XMI documents, import standard XMI documents, and import other uml models through conversion.
CWM: CWM (Common Warehouse Metamodel public Warehouse metadata model) provides a method for data format conversion, CWM can be used to describe the ing rules between two data models at any level. For example, the data entity can be transformed from a relational database into an XML format. Under the MOF framework, CWM makes it possible to transform the general data model engine.
In the OMG blueprint, UML, MOF, XMI, CWM, and other standards respectively Solve the Problems of Model Establishment, model extension, Model Exchange, and model transformation of MDA. OMG tries to expand the application scope of MDA through standardized definitions. At the same time, with such a scalable modeling language environment, IT vendors can freely implement their own modeling language and ing between languages and executable code, all must be under the OMG standardization framework.