A low-cost high-availability system realization method and device

Abstract

The invention discloses a method and a device for implementation of a low-cost high-availability system. The high-availability system adopts no independent storage device, storage resources of the high-availability system are provided by local hard disks intrinsically carried by computing resources, and the local hard disks of the computing resources in an active state and a standby state are backed up mutually in real time. When the computing resources (active nodes) in the active state are in fault, one of the computing resources in the standby state is selected according to a preset strategy to serve as a new active node and immediately take over applications and services provided for the outside by the high-availability system, and meanwhile, the storage resources are switched. By means of the scheme, the low-cost high-availability system can be implemented.

Description

technical field [0001] The invention relates to the technical field of data communication, in particular to a method and device for realizing a low-cost high-availability system. Background technique [0002] The High Availability (HA, High Availability) system effectively provides a solution to minimize the risks caused by single point failures in the network system. It improves the reliability of network system applications by shortening the downtime of network equipment due to routine maintenance operations or sudden system crashes. The HA system is currently the most effective means for enterprises to prevent the core computer system from shutting down due to failure. [0003] The core of the high availability solution is to ensure data security and service availability. Data security means that when a system failure occurs, the data itself is complete and will not be lost. When the service is restored, the data can continue to be used; service availability means that ...

AI Technical Summary

Problems solved by technology

[0010] In order to ensure the high availability of data, the design and implementation of storage resources in existing high-availability systems are relatively complicated, and the requirements for both hardware and software are extremely high

The price of a commercial storage device is often the price of dozens of servers (computing resources), which is not suitable for the application environment of small and medium-sized enterprises

[0011] 2. There is still a hidden danger of single point of failure in the data

[0012] Although the storage resources in the existing high-availability system have sufficient self-backup and protection mechanisms, due to the single physical location of the equipment, once a major accident (such as fire, earthquake, etc.) outside the equipment occurs, data security cannot be guaranteed.

[0013] 3. The synchronization complexity of storage resources and computing resources is high

[0014] Existing high-availability systems allow only one computing resource to access storage resources at a time. If two servers write data to the storage device at the same time, the data may be damaged and cannot be recovered.

In addition, due to the separation of computing resources and storage resources, the synchronization of data access is expensive, which directly affects the recovery time of failures.

[0015] 4. Waste of resources

[0016] Computing resources (servers) themselves have storage resources (server local hard disks), but business data cannot be stored on these resources, resulting in waste of resources

Images