CN104618508A - Load balancing mesh architecture based on Nginx, Memcached and Tomcat - Google Patents

Abstract

The invention discloses a load balancing mesh architecture based on Nginx, Memcached and Tomcat, belonging to the field of application program development. The architecture uses Nginx as a load balancing server and caches static resources; Memcached distributed memory object cache system as a shared synchronization session middleware; Tomcat as a web application server for deploying and running applications. Compared with the prior art, the architecture design of the present invention can distribute high-concurrency access to each application server for processing, avoiding the phenomenon that multiple services are concentrated on a certain application server and causing overload, and has good promotion and application value.

Description

A load balancing mesh architecture based on Nginx, Memcached, Tomcat

technical field

The invention relates to the field of application program development, in particular to a load balancing mesh architecture based on Nginx, Memcached and Tomcat.

Background technique

With the rapid development of informatization, the exchange of information has become more and more convenient. It can be said that as long as you have a mobile phone, you can understand the dynamics of the whole world. However, while it is convenient for people, it also brings endless pressure to the developers in the background. It is not only necessary to ensure that the functions of the application are perfect, but also to make the best and optimal design in terms of performance, so as to provide users with light, fast and safe experience.

In the existing technology, there are many ways to achieve load balancing. Different designs have different advantages and disadvantages. For example, session sharing can adopt the sticky Session design method, which is slightly better than the Memcached implementation method in terms of performance. The disadvantage is that once a certain application If a server node fails, all its corresponding services will not be able to be processed.

Contents of the invention

The technical task of the present invention is to provide a load balancing mesh architecture based on Nginx, Memcached and Tomcat aiming at the above-mentioned deficiencies in the prior art. This architecture design is aimed at the current high-concurrency access and high fault-tolerant architecture design. It can distribute high-concurrency access to various application servers for processing, and avoid the phenomenon of heavy load caused by the concentration of multiple services on a certain application server.

The technical task of the present invention is realized in the following manner: a kind of load-balanced mesh architecture based on Nginx, Memcached, Tomcat, which is characterized in that Nginx is used as a load-balanced server and caches static resources; Memcached distributed memory object caching system is used as a shared Synchronous session middleware; Tomcat is used as a web application server for deploying and running applications.

Nginx is a lightweight and high-performance web server/reverse proxy server, which is used as a load balancing server and caches static resources in the load balancing mesh architecture of the present invention;

Memcached is a high-performance distributed memory object caching system, and is used for the middleware of the synchronous session between the multiple Web server nodes in the load-balancing mesh architecture of the present invention;

Tomcat is a commonly used web application server in the prior art, and is used for deploying and running application programs in the load balancing mesh architecture of the present invention.

Preferably, at least 2 Nginx load servers are included in the gateway framework of the present invention.

The above-mentioned construction method based on Nginx, Memcached, Tomcat load balancing mesh architecture preferably includes the following steps:

1. Build a web application cluster: select several machines in the network to install tomcat application servers and deploy applications, and ensure that each application server can run normally and independently;

2. Select another machine to build the Memcached distributed memory object cache system, and associate Memcached in the tomcat application server configuration described in step 1 to realize session sharing; or use the magent cache proxy to build a Memcached cluster to realize multiple Memcached data sharing between

3. Build load server Nginx: select at least two machines to install Nginx, and configure the mapping configuration of each Tomcat in the configuration. When a user requests, the load server requests the Web application server according to the scheduling algorithm to realize the request.

The architecture design of the present invention uses multiple Nginx load balancing servers to serve multiple Web servers at the same time, a "many-to-many" model, wherein any Nginx proxy server or Web server failure will not affect the normal operation of the business, and Memcached realizes multiple Web servers. Server session sharing ensures that one of the web servers fails, and the business is smoothly handed over to other normal servers without affecting the normal business operations of users.

Compared with the prior art, the architecture design of the present invention mainly provides a solution for the high concurrent access and high fault-tolerant design of the application program, which can distribute the high concurrent access to each application server for processing, and avoid the problem caused by the concentration of multiple services on a certain application server. Phenomenon of overload.

Description of drawings

Accompanying drawing 1 is that the present invention is based on Nginx, Memcached, Tomcat load-balancing mesh structure embodiment's architecture topology diagram.

Detailed ways

With reference to the drawings in the description, taking the embodiment of three nodes as an example, the load balancing network architecture based on Nginx, Memcached, and Tomcat of the present invention will be described in detail below.

Example:

As shown in Figure 1, the architecture design includes 2 Nginx load servers, 3 Tomcat application servers and 1 Memcached. In actual operation, more nodes can be configured, and backup servers, backup load servers, cache clusters, etc. can be configured to make this architecture design more perfect.

In the above architecture design, its constituent modules can be distributed and run on different platforms. The construction method is as follows:

First of all, build a web application cluster, select three machines in the network to install the tomcat application server and deploy the application, and ensure that each application server can run normally and independently.

Choose another machine to build a Memcached distributed memory object cache system, and associate Memcached in the configuration of three tomcat application servers to realize session sharing. You can also use magent cache proxy to build a Memcached cluster to realize data sharing between multiple Memcached .

Then build the load server Nginx, select two or more machines to install Nginx, and configure the mapping configuration of each Tomcat in the configuration, so that when the user requests, the load server will request the web application server according to the scheduling algorithm to realize the request.

When some nodes in the application cluster fail, the Nginx load server will hand over the business to other normal nodes for processing, and because Memcached implements session sharing, other nodes can continue to complete business processing without any further operations.

Claims (3)

1. A load-balancing mesh architecture based on Nginx, Memcached, and Tomcat, characterized in that: Nginx is used as a load-balancing server, and caches static resources; Memcached distributed memory object caching system is used as a shared synchronization session middleware; Tomcat is used as a Web application Servers for deploying and running applications. 2. based on Nginx, Memcached, Tomcat load balancing mesh architecture according to claim 1, it is characterized in that, comprise at least 2 Nginx load servers in the architecture. 3. the method for building based on Nginx, Memcached, Tomcat load balancing mesh architecture described in claim 1 or 2, is characterized in that comprising the following steps: (1) Build a web application cluster: select several machines in the network to install tomcat application servers and deploy applications, and ensure that each application server can run normally and independently; (2) Choose another machine to build the Memcached distributed memory object caching system, and associate Memcached in the tomcat application server configuration described in step 1 to realize session sharing; or build a Memcached cluster with a magent cache proxy to realize multiple Memcached data sharing between (3) Build load server Nginx: Select at least two machines to install Nginx, and configure the mapping configuration of each Tomcat in the configuration. When a user requests, the load server requests the Web application server according to the scheduling algorithm to realize the request.