10 common software architecture patterns

Want to make APP Same thing as WeChat , Can run small programs smoothly ？ | Experience will send you to Xinjiang 、 Huawei 、 Cherry keyboard ！>>>

Python The actual combat community

Java The actual combat community

Long press to identify the QR code below , Add as needed

Scan code, pay attention to add customer service

Into the Python community ▲

Scan code, pay attention to add customer service

Into the Java community ▲

Source Java Technical station

https://mp.weixin.qq.com/s/6R4QP-gAimHzi-I8js8B6A

Want to know how to design a large enterprise class system ？ Before starting major code development , We have to choose the right Architecture , It will give us the functionality and quality attributes we need . therefore , Before we apply them to our designs , You should first understand the different architectures .

-      What is architecture pattern      -

According to Wikipedia ,

Architecture patterns are generic solutions to common problems in software architecture in a given context 、 Reusable solutions . Architecture pattern is similar to software design pattern , But more broadly .

In this paper , I will briefly introduce the following 10 Common architecture patterns , And its use 、 Advantages and disadvantages .

-      Hierarchical mode      -

This pattern can be used to build programs that can be decomposed into subtask groups , Each of them is at a specific level of abstraction . Every time we serve higher levels .
The most common in general information systems 4 The layers are divided as follows ：

• Presentation layer  The presentation layer （ That is to say UI layer ）

• Application layer  application layer （ That is the service layer ）

• Business logic layer  Business logic layer （ It's the domain level ）

• Data access layer  Data access layer （ Data persistence layer ）

application

• General desktop applications

• Electronic Commerce Web Applications

-      client - Server mode      -

The model consists of two parts ： One server and multiple clients , The server provides services to multiple clients . The client makes a request to the server , The server provides services to these clients , after , The server continues to listen for requests from clients .

application

• Online applications , E-mail 、 File sharing and banking

-      A master-slave mode      -

The model is also divided into two parts ： Master module and slave module . The master module distributes work between the same slave modules , And calculate the final result according to the structure returned from the module .

application

• In database replication , The master database is considered the authoritative data source , And keep it in sync with the database

• Peripheral devices connected to the computer system bus （ Master and slave drives ）

-      Pipeline filter mode      -

This pattern can be used to build systems that generate and process data streams . Each processing step is contained in a filter component , Data to be processed is piped . These pipes can be used to buffer or synchronize .

application

• compiler . Lexical analysis is performed using different filters in turn 、 analysis 、 Syntax analysis and code generation

• Workflow in bioinformatics

-     Broker Pattern      -

This pattern uses decoupled components to build distributed systems , These components can interact through remote service calls . The agent component is responsible for coordinating communication between components .
Servers put their functions （ Services and features, etc ） Publish to agent , Client requests service from agent , The agent then forwards the client request to the appropriate service according to its registry .

application

• Message broker software , Such as  Apache ActiveMQ, Apache Kafka, RabbitMQ  and  JBoss Messaging.

-     P2P Pattern      -

In this mode , Each individual component is called a peer （ Or peer ,peer）. A peer can act as a client （ Request services from other peers ）, It can also act as a server （ Providing services to other peers ）. The same peer may be both a client , It's the server again , And it can dynamically change its role .

application

• File sharing network , Such as Gnutella  and  G2

• Multimedia protocol , Such as P2PTV  and  PDTP

• Cryptocurrency based products , Such as bitcoin and blockchain

-      Things bus mode      -

This pattern mainly deals with components , Yes 4 An important component ： Event source 、 Event Listener 、 Channel and event bus . The event source sends the message to a specific channel on the event bus , Listeners subscribe to specific channels . When the message is sent to the channel , Listeners that subscribe to the channel receive notification of the message .

application

• Android Development

• Notification services

-     MVC Pattern      -

This pattern divides the interactive application into three parts ,

1. Model —— Including core functions and data

2. View —— Show information to users （ You can define multiple views ）

3. controller —— Handle user input

This is done to separate the internal representation of the data from the form of user input and presentation , This decouples the components , At the same time, it can also make efficient code reuse .

application

• Internet application architecture of mainstream programming languages

• Network framework , Such as Django  and  Rails.

-      Blackboard mode      -

This pattern is useful for problems where there is no definitive solution , The blackboard model consists of three parts ：

• blackboard —— A structured global memory , Contains objects in the solution domain

• Knowledge source —— Professional module with its own meaning

• Control components —— choice 、 Configuration and execution modules

All components have access to the blackboard , Components may generate new data objects to be added to the blackboard , Components look for specific types of data on the blackboard , And matching these data sources with existing patterns .

application

• speech recognition

• Vehicle identification and tracking

• Protein structure identification

• Sonar signal interpretation

-      Interpreter mode      -

This pattern is often used to design components to interpret programs written in a dedicated language , It mainly specifies how to estimate program lines , A statement or expression written in a specific language . The basic idea is to design a class for each language symbol .

application

• Database query language , Such as SQL

• The language used to describe a communication protocol

-      Architecture pattern comparison      -

Pattern	advantage	shortcoming
Hierarchical mode	A low-level service can be used by different high-level services ; Hierarchical results are easier to standardize , Because each level can be clearly defined Changes within the hierarchy do not affect other layers	It's not a pervasive Architecture ; In some cases , You need to skip some of these layers
CS Pattern	Easy to model series of services , For client request	Requests are usually responded to in different threads on the server ; Because different clients have different forms , Inter process communication can cause a lot of load
A master-slave mode	accuracy —— The execution of the service is delegated to different slave modules	The slave module is independent ： No shared state ; Communication delay between master and slave modules may be a problem , Especially in real-time systems .
Pipe filter mode	Support concurrent processing , Where input 、 When the output consists of data streams , The filter starts calculating as soon as it receives the data ; Easy to add filters , The system is easy to expand ; Filters are reusable , You can create different pipeline flows by recombining existing filters .	Overall efficiency is limited by the slowest filter program ; When passing from one filter to another , There is a load of data conversion
The proxy pattern	Allow objects to be dynamically modified 、 increase 、 Delete 、 relocation , Content distribution is transparent to developers	Service description needs to be standardized
P2P Pattern	Support decentralization operations ; It is highly stable for the failure of any node ; Highly scalable in terms of resources and computing power	There is no guarantee of service quality , Because the nodes cooperate voluntarily ; It's hard to be safe ; Performance depends on the number of nodes
Event bus mode	It's easy to add new publishers to the system 、 Subscribers and connections ; Effective for highly distributed applications	Scalability can be a problem , Because all information is transmitted through the same time bus
MVC Pattern	The first mock exam is easy to build multiple views , You can connect or disconnect at runtime	Added complexity , User actions can lead to many unnecessary updates
Blackboard mode	Easy to add new applications ; It's easy to extend the structure in the data space	It's hard to modify the structure of a data space , Because all applications will be affected ; Synchronization and access control may be required
Interpreter mode	May support highly dynamic behavior ; For end-user programmability ; Increased flexibility , Because it's easy to replace an interpreter	Because interpretive languages are usually slower than compiled languages , So performance can be a problem

Copyright notice ： Content source network , The copyright belongs to the creator . Unless you can't confirm , We all mark the author and the source , Please let me know if there is any infringement , We will delete it immediately and apologize . thank you !

 Programmer column   Scan code and pay attention to customer service   Press and hold to recognize the QR code below to enter the group 

Recent highlights are recommended ：  

  My girlfriend thinks the annual salary is 50 Ten thousand is the average level , What do I do ？

  The sexy goddess of the biggest straight men forum in China overturned

 IntelliJ IDEA Fully optimized settings , Efficiency bars ！

  Very useful Python skill

Here's a look Good articles to share with more people ↓↓