Writing to databases or mutating the underlying data in databases' is one of the important aspects of any Database Management System (DBMS). It provides a structure to your data within the underlying datastore. It is imperative to have a performance-efficient mechanism for mutation, so that your graph database can be available for further querying in the shortest possible time.
Apart from being performance-efficient, we also need to ensure that the mutation/write process follows the following principles of transaction management:
Atomicity (all or nothing): Every write is atomic in nature, so that if any part of a transaction fails, the database state is left unchanged
Consistency (from one valid state to another): At any point in time, data in the database is in a consistent state where all users read the same data
Isolation: Each transaction should occur independently and modified data should not be accessible to any other operations during a transaction...