Hi Robert and TomP, My understanding is that Robert has been working on a set of patches that would provide a separation from the clustering code that accepts transactions (front end) from the code that works to commit them (back end) and put a queue in between the two.
As I understand it, that queue could also serve to allow for pipelining in the cluster so that we could have a transactions/second rate higher than 1/latency-between-nodes. Is that right? Is this still planned for SR1? I'm just trying to help people plan their efforts for the Boron update to the performance whitepaper. --Colin
_______________________________________________ controller-dev mailing list controller-dev@lists.opendaylight.org https://lists.opendaylight.org/mailman/listinfo/controller-dev
