You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
What consistency level do you implement? For linearizable read and writes based on a quorum logic, reads must do write repair and writes must do pre-read. Your architecture docs are a bit sparse on this topic. Also, for distributed locking you need atomic-compare-and-swap which can't be implemented by a quorum based architecture as this requires consensus. As the article mentions Junodb is used for distributed locking, I wonder if the architecture description of Juno is missing important pieces.
The text was updated successfully, but these errors were encountered:
What consistency level do you implement? For linearizable read and writes based on a quorum logic, reads must do write repair and writes must do pre-read. Your architecture docs are a bit sparse on this topic. Also, for distributed locking you need atomic-compare-and-swap which can't be implemented by a quorum based architecture as this requires consensus. As the article mentions Junodb is used for distributed locking, I wonder if the architecture description of Juno is missing important pieces.
The text was updated successfully, but these errors were encountered: