High availability, two locations and three centers

1. Three centers in two places

"Two places and three centers" means the production center, the same city disaster recovery center, and the remote disaster recovery center. In a sense, it has both the same city disaster recovery and the remote disaster recovery, and the combinationIt's called the Three Centers in Two Places.

  In the same city dual centers + remote disaster recovery centers, the disaster recovery model of "two centers and three centers", the solution has both high availability and disaster backup capabilities.

1. Dual centers in the same city

Establish two data centers in the same city or adjacent cities that can independently undertake the operation of key systems. The two centers have basically equivalent business processing capabilities and synchronize data in real time through high-speed links. In daily situations, they can simultaneously share the business and management system.It can run and can be switched to run; in the event of a disaster, it can perform emergency switchover for disaster recovery without basically losing data to keep business running continuously.Compared with the remote disaster recovery model, the dual center in the same city has the advantages of low investment cost, fast construction speed, relatively simple operation and maintenance management, and higher reliability.

2. Remote Disaster Recovery Center

Establish a backup disaster recovery center in a remote city for data backup of the dual centers. Purpose: When the dual centers fail due to natural disasters and other reasons, the remote disaster recovery center can use the backup data for business recovery.

Three centers in two places: refers to the dual centers in the same city plus remote disaster recovery, a commercial disaster recovery backup solution;

Two places: refers to the same city and different places;

Three centers: refers to the production center, the same-city disaster recovery center, and the remote disaster recovery center.(production center, intra-city disaster recovery center, remote disaster recovery center)

2. Active-active data center

"Dual-active" or "multi-active" data centers are different from traditional data centers and disaster recovery centers. Multiple or two data centers are running, running the same applications and having the same data.It can provide cross-center business load balancing operation capability, realize continuous application availability and disaster backup capability, so it is called "active-active" and "multi-active"; the latter is when the production data center is put into operation, and the disaster recovery data center is not working.Only when a disaster occurs, the production data center is paralyzed, and the disaster recovery center is activated.
The biggest feature of the "active-active" data centerYes:

1. Make full use of resources to avoid waste caused by a data center being idle all the year round. Through resource integration, the service capacity of the "active-active" data center is doubled;

2. If a "active-active" data center is disconnected from one data center, its business can be quickly switched to another running data center, and the switching process is imperceptible to users.

In the "active-active" model, the two data centers accept transactions at the same time, which is very technically difficult and requires changing many underlying programs. Therefore, in reality, there is no real "active-active" data center in China. Successful application cases.

Technology selection metrics for three centers in two locations:

When building a disaster recovery system, the first consideration is to choose a reasonable data replication technology based on the actual situation.The following factors are mainly considered when choosing a reasonable data replication technology:

1. Disaster tolerance level: Identify the types of disasters that the computer system needs to endure, and the backup and protection plans for various unexpected situations such as system failures, communication failures, long-term power failures, fires and earthquakes are different.

2. Degree of business impact: It is necessary to clarify the degree of loss caused by the suspension of business when the computer system fails to work unexpectedly, that is, to define the maximum tolerance time of the user for the failure of the computer system.This is an important technical indicator for designing a backup scheme.

3. Data protection degree: Whether the database is required to restore all submitted transactions, and requires real-time synchronization to ensure data continuity and consistency, which is an important basis for the complexity of the backup scheme.