Re:In search of a simple consensus algorithm
by rescrv on Friday March 17, 2017 @11:51AM
(#13894900)
In order to get a linearizable read in Spanner, you do need to have a heartbeat from the leader of the Paxos tablet for the data you are reading in order to know that all data has been replicated up to the time at which you are taking your read. This is not free, but can be cheaper than active communication. (referred to as t_safe in the paper)
Re:In search of a simple consensus algorithm
by cube2222 on Friday March 17, 2017 @11:54AM
(#13894937)
As I understood it, if you are actually reading at some point it time (let's say 100 ms ago), then the non-leader node can check for himself, if he was in contact with the master after this point of time and give you an answer.
Re:In search of a simple consensus algorithm
by sseveran on Friday March 17, 2017 @11:43AM
(#13894819)
TrueTime is a very special case. I have not seen anyone else who has the same hardware infrastructure required to provide guarantees related to wall-clock time. Spanner also trades off transaction latency to create the wall clock time guarantees, so its not magically free.
Re:In search of a simple consensus algorithm
by cube2222 on Friday March 17, 2017 @11:50AM
(#13894892)
It isn't, I haven't said it is. The point is, there are disadvantages you are writing about, that aren't really disadvantages of the protocol itself, but disadvantages of the common implementations.
Re:In search of a simple consensus algorithm
by rystsov on Friday March 17, 2017 @01:10PM
(#13895768)
Yes that's true if all the communication between agents goes through the database then the Monotonic Reads consistency level is equivalent to linearizability.
In case there are off the database communications then this level of consistency isn't enough.
(Not talking about etcd, check out the spanner paper)