Redundancy

What is Redundancy?

Redundancy refers to a method of constructing a system by preparing multiple components (such as devices or communication lines) with identical functionalities and combining them to form the overall system. This approach ensures that even if some components fail or go offline, the system can continue operating, thereby improving availability. Depending on the configuration, redundancy can also enhance performance.

Redundant configurations are broadly categorized into two types: HA (High Availability) clusters and LB (Load Balance) clusters:

• HA Clusters: These pass on the "state" of failed system components to backup components during failover. This configuration is typically used for systems requiring data integrity, such as database servers, with the primary goal of ensuring high availability.

  • In an Active/Active configuration, multiple components with the same functionality operate simultaneously, and if one fails, the others continue processing.
  • In an Active/Standby configuration, some components are active while others remain on standby. If an active component fails, a standby component takes over. The N+1 configuration, where N components are active and 1 is on standby, is a variation of the Active/Standby setup.

   

• LB Clusters: These do not transfer the "state" between components during failover. Commonly used for web servers and application servers, LB clusters focus on performance improvements through load balancing while also contributing to higher availability.