Hello,
Forgive me if I have missed anything in the obvious locations, but I am trying
to find out if anyone has done an analysis of the gossip protocol as
implemented in Cassandra? In particular, I am interested in the the theoretical
propagation time (to all nodes) of a change. For example,
There is this very old ticket:
https://issues.apache.org/jira/browse/CASSANDRA-617
But note that is gossip simulation, not the actual gossiper Cassandra uses
(and also very antiquated.) Using the actual gossiper under simulation is
unfortunately complicated by
Thanks for this info, I will read it now.
The idea is, I have a modified version of Cassandra for a specific
research application. I want to be able to make a supportable statement
like this:
After a state change, we wait x*gossip_interval in order to ensure with
0.9 probability that