Analysis of replication Master-slave architecture based on Mysql database _mysql

In order to apply the scalability of the system, often need to scale out the database design, scale out design is to increase the database processing node to improve the overall processing capacity of the system, that is, increase the number of database servers to share the pressure. In this way the scalability of the system is enhanced, the cost is reduced, but the system architecture is complex and maintenance is difficult. It is unavoidable that the system's downtime or failure occurs. Thus, in theory, the security of the system (possibly data loss) is reduced and availability is reduced. So to improve data security, as well as the high availability of the system, the simple way is that all hardware and software to avoid a single point of vulnerability, all the data are stored in multiple copies. Technically, it can be achieved through the database replication technology. The replication technology of MySQL is one of the means to realize database replication.

With regard to MySQL's replication, there are several alternative architectural scenarios, such as the General replication Architecture Master-slave scenario, the Dual Master architecture scenario, and the Cascade replication architecture scenario, focusing today on the simplest master-slave scenarios, Its schema diagram looks like this:


This architecture scenario, which is a MySQL server as the primary server, is mainly responsible for the application of the client's write data processing, while others for the MySQL server as slave to replicate the data source. And those slave server data is a full backup of master server data, relatively real-time backup, these backup data mainly supplied with the client read data used, because the general application system read data pressure than write data, especially Web application system.