Shulou(Shulou.com)05/31 Report--

This article shares with you the content of the sample analysis of Spring Cloud. The editor thinks it is very practical, so share it with you as a reference and follow the editor to have a look.

1 Spring Cloud Overview 1.1 traditional applications 1.1.1 monomer applications

Prior to this, the author's company developed Java programs, mostly using Struts, Spring, Hibernate (MyBatis) and other technical frameworks, and each project will release a single application. For example, to develop a purchase, sale and storage system, a war package will be developed and deployed to Tomcat. Every time you need to develop a new module or add a new function, it will continue to be added on the original basis. Several years later, the war package continues to expand, programmers in debugging, the server may also need to start for half a day, the maintenance of the system is extremely inefficient. Such a war package covers inventory, sales, membership, reporting and other modules, as shown in figure 1-1.

Figure 1-1 monomer application

There are many hidden dangers in such single applications, and any bug may lead to the downtime of the whole system. What impressed me most was that there was a customer who exported a sales detail report (with a large amount of data) during the peak period, which eventually paralyzed the whole system and unable to sell the sales staff at the front desk. Maintaining such a system is not only extremely inefficient, but also full of risks. All members of the project team are in a panic, and we need a fundamental change.

1.1.2 Architectural evolution

In order to solve the problems of the above single applications, we refer to the SOA architecture, divide each module into independent service modules (war), and use the read-write separation of the database, as shown in figure 1-2.

Figure 1-2 Architectural evolution

Each module will call each other, for example, the sales module will call the interface of the member module. In order to reduce the coupling between each module, we have added the Enterprise Service bus (ESB). The architecture between each module and ESB is shown in figure 1-3.

Figure 1-3 ESB

After joining ESB, each module publishes services to ESB, and they communicate with ESB using SOAP protocol. After the implementation of the architecture of figure 1-2 and figure 1-3, the performance of the whole system has been significantly improved, and the coupling of each module has been reduced. After running for a period of time, there is a new problem. Due to the increase in the number of sales terminals, the sales module obviously exceeds its capacity. In order to ensure the normal operation of the sales front end, we use Nginx to do load balancing, as shown in figure 1-4.

Figure 1-4 using Nginx

With the increase of sales modules, there are many problems, such as managing these modules, which is an arduous task for operation and maintenance engineers. Once the sales modules are modified, they will upgrade all night. In addition, the enterprise service bus may also become a performance bottleneck, although this problem has not occurred yet, but we need to plan ahead.

1.1.3 Architectural requirements

From the previous architecture evolution, we can see that every point in the application may become a problem point of the system. With the popularity of Internet applications, in the environment of big data and high concurrency, our system architecture needs to face more stringent challenges. We need a new architecture, which can at least meet the following requirements:

 high performance: this is the basic requirement of the application.

 independence: one of the modules has bug or other problems that cannot affect other modules or the entire application.

 is easy to expand: every node in the application can be extended according to the actual needs.

 is easy to manage: the resources of each module can be easily managed and upgraded to reduce maintenance costs.

 status monitoring and alarm: monitor the entire application and issue an alarm in time when there is a problem with a node.

In order to solve the problems encountered and meet the above architectural requirements, we began to study Spring Cloud.

1.2 what is a micro service and a Spring Cloud1.2.1?

The word micro-service comes from the article "Microservices" of Martin Fowler. Micro-service is an architectural style, which divides individual applications into small service units, and uses HTTP's API to access and operate resources between micro-services.

In the division of single applications, micro-services are somewhat similar to the previous SOA architecture, but the SOA architecture focuses on integrating the services of each single application into ESB, while micro-services are done more thoroughly, emphasizing that by turning the whole module into service components, micro-services may partition modules with finer granularity. Take our previous sales and membership modules as an example, in the SOA architecture, you only need to publish the corresponding services to the ESB container, while in the micro-service architecture, the two modules themselves will become one or more service components. The SOA architecture and the microservice architecture are shown in figures 1-5 and 1-6.

Figure 1-5 SOA architecture

Figure 1-6 Micro-service architecture

In terms of the architecture of microservices, Martin Fowler's article affirms the contribution of Netflix. Next, let's take a look at Netflix OSS.

1.2.2 about Netflix OSS

Netflix is an Internet film provider. A few years ago, Netflix set up its own open source center, called Netflix Open Source Software Center, or Netflix OSS for short. This open source organization focuses on big data and cloud computing technology, and provides a number of open source frameworks, including big data tools, construction tools, cloud-based service tools, and so on. These frameworks provided by Netflix well follow the concept advocated by micro-services, and achieve decentralized service management, service fault tolerance and other mechanisms.

1.2.3 Spring Cloud and Netflix

Spring Cloud is not a specific framework, we can understand it as a toolbox, it provides a variety of tools that can help us to quickly build distributed systems.

Each project of Spring Cloud is based on Spring Boot, encapsulates multiple frameworks of Netflix, and binds these frameworks to the environment of Spring through automatic configuration, thus simplifying the use of these frameworks. Because of the simplicity of Spring Boot, it is easy for us to integrate Netflix frameworks into our project when using Spring Cloud. The "Spring Cloud Netflix" module under Spring Cloud mainly encapsulates the following items of Netflix:

 Eureka: distributed middleware based on REST services, mainly used for service management.

 Hystrix: fault-tolerant framework that helps us control the interaction of components between distributed systems by adding latency thresholds and fault-tolerant logic.

 Feign: a REST client designed to simplify the development of Web Service clients

 Ribbon: a load balancing framework that supports communication between clients in a micro-service cluster. It mainly implements load balancing for middle-tier applications.

 Zuul: provides proxy, filtering, routing and other functions for micro-service clusters.

1.2.4 main modules of Spring Cloud

In addition to the Spring Cloud Netflix module, Spring Cloud includes the following important modules:

 Spring Cloud Config: provides configuration servers and configuration clients for distributed systems, through which configuration files in the cluster can be well managed.

 Spring Cloud Sleuth: service tracking framework, which can be integrated with data analysis and service tracking systems such as Zipkin, Apache HTrace and ELK to facilitate service tracking and problem solving.

 Spring Cloud Stream: a framework for building message-driven microservices that integrates "Spring Integration" to connect message broker middleware on the basis of Spring Boot.

 Spring Cloud Bus: a cluster message bus that connects message agents such as RabbitMQ, Kafka, etc.

Thank you for reading! This is the end of this article on "sample Analysis of Spring Cloud". I hope the above content can be of some help to you, so that you can learn more knowledge. if you think the article is good, you can share it for more people to see!

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.