As mentioned in the article, a non-100%-correct lock can be used for efficiency purposes. So basically use an imperfect locking mechanism for efficiency and a reliable one for correctness.
To be clear, my point is don't use distributed locking for correctness. There are much better options.
Now, the atomicity I mention implies some kind of internal synchronization mechanism for multiple requests, which could be based on locks, but those would be real, non-distributed ones.
To be clear, my point is don't use distributed locking for correctness. There are much better options.
Now, the atomicity I mention implies some kind of internal synchronization mechanism for multiple requests, which could be based on locks, but those would be real, non-distributed ones.