It seems to me that the per-record timeline consistency as defined by Cooper et al. in "PNUTS: Yahoo!'s Hosted Data Serving Platform" mimics the (older?) definition of monotonic writes. From the paper:
per-record timeline consistency: all replicas of a given record apply all updates to the record in the same order.
This is quite similar to a definition for monotonic writes:
A write operation by a process on data item x is completed before any successive write operation on x by the same process.
Can I conclude that those things are the same, or is there a difference that I misunderstand? Note that the link above also mentions possible copies of data item x, so monotonic write includes replicas.
Asked By : evnu
Answered By : Jan
I'm not sure these two concepts are really comparable.
monotonic writes is a client based consistency model. It defines what a process sees.
per-record timeline consistency strikes me as a memory based model. It doesn't even mention what a client sees or does.
So I would say they are not same or even comparable.
Best Answer from StackOverflow
Question Source : http://cs.stackexchange.com/questions/2339
0 comments:
Post a Comment
Let us know your responses and feedback