Consistency is a challenge when having concurrent systems/users.
Having multiple versions of the same value at more than one place, but not having all the versions updated.
The duplicated data, also known as the redundant data, creates unreliable information.
= no inconsistency
A read is guaranteed to return the most recent write for a given client.
Assuming you have a storage system with more than one machine, consistency implies that the data is the same across the cluster, so you can read or write to/from any node and get the same data.
= temporarily inconsistent, it takes time to propagate.
Exactly what the name suggests. If multiple machines store the same data in a cluster, an eventually consistent model implies that all machines will eventually have the same data.
It’s possible that at a given time, machines have different versions of the same data (temporarily inconsistent), but they will eventually reach a state where they have the same data.