Wondering if you are going to deal with a problem with the right solution?

Redundancy: the ability to maintain multiple identical copies of a service component, and to use those copies in a way that prevents any one component from becoming a single point of failure.

Replication: The ability to maintain multiple copies of data, called replicas

Backup: The ability to maintain a timestamped copy of data that can be used to restore data that has been lost

At first glance, these definitions may look very similar, if not the same. For sure, they are overlapping concepts, but they are expected to handle different issues.

We could simplify by saying that redundancy is a sort of backup system to avoid downtimes. While replication is a sort of data backup to avoid data loss. And backup itself, is what can be used in case an incident happens, that requires to recover data.

Looking more in detail, redundancy consists in deploying multiple instances of a component. It is not so difficult to apply to stateless resources, but when involves stateful ones, it can quickly become complex to handle. It can be used to prevent downtime, enabled in case of failure or regional outage. And can be combined with a load balancer, to handle traffic across redundant systems, improving availability.

Redundancy applied to data, is called replication. The idea, is to maintain multiple real-time or near real-time copies of live data, called replica. To improve resiliency, it is possible to distribute replicas over different locations.
Despite the approximation done before, replication is not the same as backup, because replication synchronize all changes among multiple replicas, without keeping old copies of data. It can be used to provide high availability of data, even in case of failures. To improve read performances, by allowing distributed access. To enable data for global applications.

Finally, a backup takes a copy of the data at a specific point in time, and multiple copies are often stored, to select one to be restored in case of problems. Backup, are used in case of catastrophic losses, data corruption or deletion, cyberattacks.

Conclusion

Different cloud providers, may have a slightly different definition. But concepts are essentially the same. All these concepts spin around having multiple instances, but in one case, redundancy, it is applied to a component or a system. Replication is about having multiple instances of live data, all available to be accessed, that require an high degree of synchronization to work as intended. ANd finally there is the backup, that are versioned copies, normally stored aside, to be used only in case there is a big issue that require to get data from old archives.