The data might not be consistent immediately but eventually, it becomes consistent

The state of the system is always soft which means that it can change over time even if there are no reads or writes as the system keeps changing the data to make it consistent

when a database action is committed it forms the permanent state of database

parts of a transaction shouldn't commit if any one of them fails

any transaction should lead database from one stabe state to another

The system works well despite physical network partitions

true consistency cannot be achieved in a distributed environment

All clients always have the same view of the data

in a distributed system, a maximum of two of the three properties availability, consistency, partition tolerance can be guaranteed

Reads in the system are still possible even though they may not give the correct response due to inconsistency

Each client can always read and write

Either the whole transaction should succeed or it should be complete rollback