What is a layered architecture model?

The layer mode is the most common architecture mode. Even the layered mode is the basis of many architectural models. Some of the content mentioned in this chapter is actually related to the layered mode.

The hierarchy describes the architectural design process starting with the lowest level abstraction, known as layer 1st. This is the basis of the system. Step by step up the abstraction tier by placing the tier J above the tier J-1 until it reaches the highest level of functionality, called the tier n.

Therefore, the layered mode can be defined as separating the components of the solution into different layers. Components in each layer should be cohesive and at the same abstract level. Each layer should be loosely coupled with the layers below it.

 

The key point of the layered mode is to determine the dependency: By layering, the dependencies between subsystems can be restricted to make the system more loosely coupled and easier to maintain.

 

Another concept is partition. Hierarchy refers to the horizontal division of the architecture, while partitioning refers to the vertical division of the architecture.

The typical layering method is that the application-specific features are located at the upper layer, the features that span the application domain are located at the middle layer, and the configuration environment-specific features are located at the lower layer. The number and composition of layers depend on the complexity of the problem areas and solutions. Generally, there is only one application dedicated layer. The sub-system should be organized into a component layer structure. The upper layer of the architecture is an application-specific sub-system, the lower layer is a hardware and operation-specific sub-system, and the middleware layer is a General Service.

 

The basic principles for system layering are as follows:

-Visibility. Subsystems must be dependent on subsystems at the same layer and at the next layer.

-Easy to change. At the top layer, elements that change as user needs change. Underlying elements that change with the changes in the implementation platform (hardware, language, operating system, database, etc. Intermediate interlayer placement is widely used in various system and implementation environments. If further division in these categories helps to organize the model, add more layers.

-Versatility. Abstract model elements are generally placed at the lower layer of the model. If they are not specific to the specific implementation, they tend to be placed in the middle layer.

-Layers. For small systems, three layers are enough. For complex systems, layer 5-7 is usually required. Regardless of the complexity, if there are more than 10 layers, you need to think carefully. The more layers, the more careful you need.

Common layered architecture models
1. client-server model (client-server, C/S ).

2. layer-3 model: User presentation layer, business logic layer, and data layer.

3. Multi-layer structure technical composition model: presentation layer, middle layer, data layer.

4. Three Layers of network system structure: core layer, aggregation layer, and access layer.

5. Typical layers of RUP: application layer, professional service layer, middleware layer, and system software layer.

6. B/S Mode System Structure Based on Java: browser, server, request receiving layer, and request processing layer.

7. A six-layer structure: function layer (User Interface), Module layer, assembly layer (Software Bus), service layer (data processing), data layer, and core layer.

The layer mode is the most common architecture mode. Even the layered mode is the basis of many architectural models. Some of the content mentioned in this chapter is actually related to the layered mode.

The hierarchy describes the architectural design process starting with the lowest level abstraction, known as layer 1st. This is the basis of the system. Step by step up the abstraction tier by placing the tier J above the tier J-1 until it reaches the highest level of functionality, called the tier n.

Therefore, the layered mode can be defined as separating the components of the solution into different layers. Components in each layer should be cohesive and at the same abstract level. Each layer should be loosely coupled with the layers below it.

 

The key point of the layered mode is to determine the dependency: By layering, the dependencies between subsystems can be restricted to make the system more loosely coupled and easier to maintain.

 

Another concept is partition. Hierarchy refers to the horizontal division of the architecture, while partitioning refers to the vertical division of the architecture.

The typical layering method is that the application-specific features are located at the upper layer, the features that span the application domain are located at the middle layer, and the configuration environment-specific features are located at the lower layer. The number and composition of layers depend on the complexity of the problem areas and solutions. Generally, there is only one application dedicated layer. The sub-system should be organized into a component layer structure. The upper layer of the architecture is an application-specific sub-system, the lower layer is a hardware and operation-specific sub-system, and the middleware layer is a General Service.

 

The basic principles for system layering are as follows:

-Visibility. Subsystems must be dependent on subsystems at the same layer and at the next layer.

-Easy to change. At the top layer, elements that change as user needs change. Underlying elements that change with the changes in the implementation platform (hardware, language, operating system, database, etc. Intermediate interlayer placement is widely used in various system and implementation environments. If further division in these categories helps to organize the model, add more layers.

-Versatility. Abstract model elements are generally placed at the lower layer of the model. If they are not specific to the specific implementation, they tend to be placed in the middle layer.

-Layers. For small systems, three layers are enough. For complex systems, layer 5-7 is usually required. Regardless of the complexity, if there are more than 10 layers, you need to think carefully. The more layers, the more careful you need.

Common layered architecture models
1. client-server model (client-server, C/S ).

2. layer-3 model: User presentation layer, business logic layer, and data layer.

3. Multi-layer structure technical composition model: presentation layer, middle layer, data layer.

4. Three Layers of network system structure: core layer, aggregation layer, and access layer.

5. Typical layers of RUP: application layer, professional service layer, middleware layer, and system software layer.

6. B/S Mode System Structure Based on Java: browser, server, request receiving layer, and request processing layer.

7. A six-layer structure: function layer (User Interface), Module layer, assembly layer (Software Bus), service layer (data processing), data layer, and core layer.