Yes, VLL algorithm kinda solves the throughput problem. You will still have some issues with fault-taulerance - i.e. what you do if a machine does not answer yo

Yes, VLL algorithm kinda solves the throughput problem. You will still have some issues with fault-taulerance - i.e. what you do if a machine does not answer you. For intra-process framework it's not an issue (as long as we do not have deadlocks).