CN104468747A - High-performance deployment method based on B/S - Google Patents

Abstract

The present invention relates to the technical field of computer software development and deployment, and in particular to a high-performance deployment method based on B/S. First, a database cluster is deployed, and the corresponding data between clusters can be modified by triggers; each database corresponds to an application server and only serves one application server; multiple application servers establish a cluster, which is uniformly debugged by the front-end scheduling server; the scheduling server sends the request to the relatively idle server for response according to the status of the server in the cluster, so as to achieve the effect of load balancing; the data access each other using triggers with higher efficiency to ensure the data consistency of the entire platform. The present invention meets the high concurrency requirements of the platform and can be used for application deployment.

Description

A high-performance deployment method based on B/S

technical field

The invention relates to the technical field of computer software development and deployment, in particular to a B/S-based high-performance deployment method.

Background technique

At present, in terms of application deployment, many cases are completed by using an application server plus a database server; and in some good deployment solutions, application service clusters will be built, but in terms of database deployment, in order to make data consistent and effective, By deploying only one database server. It is equivalent to one database server serving multiple application servers. Therefore, in the case of high concurrency, the bottleneck of the database is very obvious. The performance of the application server is improved, but the overall performance is very poor, and even the database server may go down.

Contents of the invention

The technical problem solved by the present invention is to provide a B/S-based high-performance deployment method to meet the high concurrency requirements of the platform.

The technical scheme that the present invention solves the problems of the technologies described above is:

First deploy a database cluster, and the clusters can modify the corresponding data through triggers; each database corresponds to an application server and serves only one application server; multiple application servers establish a cluster, and the front-end scheduling server is unified Debugging; the scheduling server sends requests to idle servers to respond according to the status of the servers in the cluster, achieving the effect of load balancing; the data uses highly efficient triggers to access each other to ensure the data consistency of the entire platform.

Described method comprises following specific steps:

The first step is to establish a database cluster, and the databases in the cluster can access and modify data through triggers;

The second step is to establish an application cluster, and the clusters are independent of each other; at the same time, one application server corresponds to one database server to avoid database sharing;

Step 3: Deploy the scheduling server, which can distribute the request to any application server in the cluster;

Step 4, the user accesses the scheduling server through a browser;

Step 5: The scheduling server sends the request to a certain server A for response according to the status of each server in the cluster;

Step 6, the server accesses the corresponding database according to the business logic;

Step 7, the database judges whether it is a query service, if yes, directly query data from server D, return it, and skip to step 9; otherwise, the process skips to step 8;

Step 8, database server A modifies the data of other database servers in the cluster through triggers;

Step 9, end.

The beneficial effects of the invention are: the method solves the database bottleneck problem under high concurrency by deploying database clusters, and the clusters can modify each other through triggers, and at the same time, each database is only accessed by one application server. At the same time, when the data changes, the databases modify the data of other databases in the cluster through triggers to achieve the effect of data consistency. On the front-end application server, the request is forwarded to an idle server through the scheduling service. Therefore, this method can provide a highly concurrent application deployment method on the premise of ensuring data validity and consistency.

Description of drawings

Below in conjunction with accompanying drawing, the present invention is further described:

Fig. 1 is a flow chart of the present invention;

Fig. 2 is a flowchart of a specific deployment case of the present invention.

Detailed ways

The system of the invention provides high-performance concurrent services on the premise of achieving data validity and consistency by establishing a database cluster and a data modification mechanism.

As shown in Figure 1, the basic flowchart of the present invention is:

The first step is to establish a database cluster, and the databases in the cluster can access and modify data through triggers;

The second step is to establish an application cluster, and the clusters are independent of each other; at the same time, one application server corresponds to one database server to avoid database sharing;

Step 3: Deploy the scheduling server, which can distribute the request to any application server in the cluster;

Step 4, the user accesses the scheduling server through a browser;

Step 5: The scheduling server sends the request to a certain server A for response according to the status of each server in the cluster;

Step 6, the server accesses the corresponding database according to the business logic;

Step 7, the database table, according to whether the data has been modified, modify the data of other database servers in the cluster through the trigger;

Step 8. Server A returns the corresponding response to the user.

Taking the file uploading of the education cloud system as an example, and two database servers and two application servers as examples, the main deployment of the present invention will be described below.

As shown in Figure 2, it is a specific deployment case flow of the present invention; it includes the following steps:

Step 1, deploy database server DA and database server DB;

Step 2. Configure the file table T1 of DA. When new data is added to T1, the data is added to T1 of DB through a trigger; at the same time, table T1 of DB is configured;

Step 3. Deploy the application server WA and WB, and associate WA with DA, and associate WB with DB;

Step 4, deploy scheduling service R;

Step 5, the user uploads the file through R.

Step 6, R sends the user request to the idle application server WA at that time;

Step 7, WA judges whether it is a query business, if yes, returns from DA query and ends, otherwise the process skips to step 8

Step 8, WA adds data to table T1 of DA.

Step 9: T1 of DA writes data into table T1 of DB through a trigger;

Step 10, end.

Claims (2)

1. A high-performance deployment method based on B/S, which is characterized in that: first deploy a database cluster, and the corresponding data can be modified each other through triggers between the clusters; each database corresponds to an application server, and only for one application Server service; multiple application servers establish a cluster, which is uniformly debugged by the front-end scheduling server; the scheduling server sends requests to idle servers to respond according to the status of the servers in the cluster, achieving the effect of load balancing; data usage efficiency Higher triggers access each other to ensure data consistency across the platform. 2. The high-performance deployment method based on B/S according to claim 1, characterized in that: said method comprises the following specific steps: The first step is to establish a database cluster, and the databases in the cluster can access and modify data through triggers; The second step is to establish an application cluster, and the clusters are independent of each other; at the same time, one application server corresponds to one database server to avoid database sharing; Step 3: Deploy the scheduling server, which can distribute the request to any application server in the cluster; Step 4, the user accesses the scheduling server through a browser; Step 5: The scheduling server sends the request to a certain server A for response according to the status of each server in the cluster; Step 6, the server accesses the corresponding database according to the business logic; Step 7, the database judges whether it is a query service, if yes, directly query data from server D, return it, and skip to step 9; otherwise, the process skips to step 8; Step 8, database server A modifies the data of other database servers in the cluster through triggers; Step 9, end.