Shulou(Shulou.com)06/02 Report--

This article shows you what the distributed structure of Java enterprise application architecture design is like, the content is concise and easy to understand, it can definitely brighten your eyes. I hope you can get something through the detailed introduction of this article.

The distributed structure in the design of Java enterprise application architecture can be divided into single-level structure, 2-level structure, 3-level structure and N-level structure. Full understanding and application of distributed architecture can better understand the current situation of contemporary network computing and design better enterprise applications.

For a long time, Java Enterprise Edition (Java EE) has become the platform of choice for enterprise business application development and deployment in many industries (such as banking, insurance, retail, hotel, tourism, telecommunications, etc.). The reason why Java EE is so widely used is that Java EE can provide a standardized platform for building robust and highly scalable distributed applications, which can range from core business operations of banks to airline booking engines. However, the successful development of Java EE applications can also become a difficult task, in which Java enterprise application architecture design plays an important role.

First of all, the rich choices offered by the Java EE platform itself can be daunting. The plethora of frameworks, utility class libraries, integrated development environments (IDE), and alternative tools make everything more challenging. Therefore, choosing the right technology is very important for the development of Java EE-based software. Those technologies with robust architecture and design principles will be of great benefit to building applications that are easy to maintain, reuse and extend.

We will first review the evolutionary history and n-level structure of distributed computing. After that, I will show how the Java EE platform solves the difficulties in distributed application development. You will also learn about model-view-controller (MVC) structure guidelines. Then I will combine the MVC guidelines with the Java EE platform to explain the multi-tier Java EE application structure.

After understanding the application system architecture, I will focus on the development of Java EE applications based on object-oriented principles. I will also explain how to use design patterns to simplify the design process and how to choose practical examples of *. In addition, I will touch on the catalogue of design patterns included in Sun's Java BluePrints, which is described in detail in Deepak Alur et al's Core J2EE Design patterns (Prentice Hall Press, 2003). In this article, I will introduce the General Modeling language (UML) and its role in visual Java EE document design and architecture.

Evolutionary History of distributed Computing

In distributed computing, an application is divided into several smaller components and runs on different computers at the same time. This method of computing is also known as "network computing" because these components usually communicate through some protocols based on TCP/IP or UDP protocols. These smaller applications are called "levels", and each level can independently provide a class of services to other connection levels. The "level" can be subdivided into several "layers" in order to reduce the granularity of the function. Most Java enterprise application architecture designs should have three different layers:

The ◆ presentation layer is responsible for the user interface.

The ◆ business layer executes business logic. During operation, it also interacts with the data access layer.

The ◆ data access layer is responsible for operations such as accessing the data stored in the enterprise information system (EIS).

By analyzing the transition history of distributed computing structure, we can better understand the current situation of contemporary network computing. In the next few sections, I will use a few appropriate examples to introduce the evolution of distributed architecture.

Single-stage structure

The use of single-stage structures can be traced back to the days when giant hosts were connected with simple terminals. In this structure, all application layers, such as user interface, business logic, and data, are configured in the same physical host. The user interacts with the system through the terminal or console, which has only very limited text processing power (see figure 1)

Figure 1. Single-tier structure (text in the picture: Console-- "console"; Dumb Terminal-- "simple terminal"; Mainframe-- host)

Two-level structure

In the early 1980s, personal computers (PC) became very popular, cheaper than mainframes and more powerful than devices such as simple terminals. The advent of PC paves the way for truly distributed (client-server, Cpicurus S) computing. PC, as a client, can now run client interface (UI) programs independently, while it also supports graphical client interface (GUI), which allows users to enter data and interact with the server host, which is now only responsible for the business logic and data. When the user completes the data entry in the client, the GUI program can selectively check the validity of the data, and then send the data to the server for business logic processing. The form-based application of Oracle is an excellent example of a two-tier structure. The GUI of the form is stored in the client-side PC, while the business logic (including code and stored procedures) and data remain in the database server of Oracle.

Since then, another level 2 structure has emerged, in which not only the user interface (UI), but also the business logic is placed at the client level. The typical operation mode of this application is to connect directly to the database server for various database queries. This client is called a "fat client" because this structure puts a significant portion of the executable code at the client level (see figure 2).

Figure 2. Level 2 structure (Business Logic Layer-- business logic layer; Optional-- optional; User Interface Layer-- user interface layer; Thick Client-- rich client; Data Access Layer-- data access layer; Mainframe Server-- server host)

Three-level structure

Although the development of level 2 "rich client" applications is simple, any software upgrade caused by a change in user interface or business logic needs to be done on all clients. Fortunately, in the mid-1990s, the cost of hardware has become lower and lower, while CPU's computing power has been greatly improved. At the same time, the development of the Internet is very rapid, the development trend of Internet applications has gradually emerged, the combination of the two finally led to the emergence of the three-level structure.

In the three-tier structure model, PC clients only need to install "thin client" software, such as browsers, to display the display provided by the server, the server is responsible for preparing the display content, business logic, and data access logic, and the application's data comes from enterprise information systems, such as relational databases. In such a system, business logic can be accessed remotely, so supporting a stand-alone client through a Java console application becomes a course. The business layer mainly interacts with the information system through the data access layer. Because the entire application is on top of the server, such a server is also called "application server" or "middleware" (see figure 3).

Figure 3. 3-level structure (text in figure: Presentation Layer-- presentation layer; Business Logic Layer-- business logic layer; Data Access Layer-- data access layer; Thin Client-- thin client; Application Server-- application server; Enterprise Data-- enterprise data; Database Server-- database server)

N-level structure

With the continuous improvement of Internet bandwidth, major enterprises around the world have launched their network services one after another. As a result of this change, the application server can no longer bear the heavy load of the presentation layer. This task has now been undertaken by a dedicated web server that is specifically responsible for generating display content. The content is then transferred to a client-level browser, which is responsible for displaying the user interface. The application server in the N-level structure is responsible for providing business logic components that can be accessed remotely, while the presentation layer web server accesses these components through the network using the local network protocol. Figure 4 shows the n-level structure.

The above are the distributed structures in the design of Java enterprise application architecture. in different requirements and application scenarios, we will use different distributed structures and design different Java enterprise application architectures.

The above is what the distributed architecture in Java enterprise application architecture design is like. Have you learned the knowledge or skills? If you want to learn more skills or enrich your knowledge reserve, you are welcome to follow the industry information channel.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.