Clustering (Active Clustering, Standby Clustering)

152 단어·1 분·원문(.md)

Clustering #

One of the database distribution techniques that involves having multiple DB servers to prepare for when one server goes down.

Active Clustring #

Multiple DB servers are configured, and each server is kept in an Active state.

Advantages #

  • Even if one server goes down, other servers immediately take over its role, so there is no service interruption.
  • CPU and memory utilization can be increased.

Disadvantages #

  • If a single storage is shared, a bottleneck may occur.
  • Operating multiple servers simultaneously incurs higher costs.

Standby Clustering #

Only one server is operated, and the remaining servers are kept in a Standby state. If the operating server goes down, the Standby server is switched to an Active state.

Advantages #

  • It costs less compared to Active-Active clustering.

Disadvantages #

  • When a server goes down, time is required to switch the Standby server to an Active state.
DataBase/clustering.md