What is the difference between Mutator and Shared Mutator?
In the specs for Mutator, there is a "flush_interval" (Auto-flush interval in milliseconds; 0 disables it.) parameter.
In the specs for Shared Mutator, it states:
WARNING: Shared mutators sacrifice write durability guarantee because they buffer and flush mutations on a periodic time interval and the set_ methods return without knowing if the mutations succeeded or failed.
So, what is the difference if both have flush intervals?
Further, are both thread-safe (in Java)?
Also, which is more efficient in terms of performance?
It seems to me that, because the Shared Mutators do not check on the return atatement, that they would run faster and potentially allow several parallel actions. This comes at a trade-off, providing that it does not guarantee success.
The mutator seems to be more reliable, but it would take more time (with a small file size, the difference would be negligable).
I will expand this when I have more time