Glossary · System coordination, integration and orchestration
Active-active redundancy
Also known as: hot-hot redundancy, load-sharing redundancy
German: Aktiv-Aktiv-Redundanz
In system architecture, active-active redundancy is a redundancy scheme in which two or more redundant nodes run and serve requests at the same time. If one node fails, the remaining nodes carry the load without a switchover to a standby unit.
- System integration
In one sentence
In active-active redundancy all redundant nodes work at the same time, so the loss of one node leaves the others serving without a switchover.
Example
Two MES application servers behind a load balancer both process work-order requests; when one server is patched, the other handles all traffic.
How it applies
- Engineering: Active-active designs need shared or replicated state, so Data replication and the chosen Consistency model become part of the architecture. Each node must be sized to carry the full load if its partner fails.
- Operation: Monitor that all nodes are really active; a node that silently stopped serving turns the system into a single point of failure without anyone noticing.
- Documentation: Describe in the system manual how many nodes may fail before service degrades, how the operator sees the redundancy status and how a failed node is returned to service.
Active-active vs. active-standby redundancy
In Active-standby redundancy one unit works while the other waits and takes over only on failure. Active-active uses all capacity all the time and avoids a switchover delay, but it is harder to keep data consistent. Neither scheme protects against a Common-cause failure (CCF) that hits all nodes at once, such as a shared software defect.