> 3) (hopefully) a lot of it is coming from the fact that quite a few DC > operators believe that full transactional semantics that can be used by > some form of a guaranteed consistency model is absolutely mandatory for > this mapping system. Since as far as I know, that's not something BGP has, > then it is simply not the right tool for the job, and any number of > distributed database/state synchronization methods are a significantly > better fit, and should be used instead..
Yes, storing the tunnel inner to outer relationships and the sets of other members of a given L2 domain should be a cake-walk for something like Hadoop/HDFS and there is a list of about 30 odd DC's already using it on impressive scales so clearly that would work. However the challenge here is that Hadoop and others of its ilk are not standardized at the IETF so procedurally how would that work? Or are we going to have to get into BGP to Hadoop interworking ;) Peter _______________________________________________ nvo3 mailing list [email protected] https://www.ietf.org/mailman/listinfo/nvo3
