Hi Robert,
> > then TTE would simply shift load around and drop things elsewhere. > > Yes exactly. Is this good ? No, it’s the same: congestion and packet loss. It is not global network meltdown. > > TTE cannot create bandwidth. > > I think that is pretty obvious to everyone. Not sure why you keep repeating > it in each mail :) You continue to criticize TTE because it does not somehow miraculously solve a situation where the network is uniformly super-saturated. Those situations are hardly typical. > The point is that during TTE provisioning (enabling TTE on a per node or per > interface of the node basis) alternative links may not be congested. But > during unexpected load there can be congestion on backup paths and nodes > executing TTE will have zero knowledge about it. They will simply redirect > part of the traffic only hoping that other paths perhaps are idle enough. That’s correct. If they are, then TTE has helped. If there is no backup capacity then TTE won’t be effective. > > There is nothing that can, except more bandwidth and I see no benefit to > > even discussing this situation. > > Well to start with you are assuming that end to end TE is very badly > provisioned as if it would be built correctly all available paths can be > utilized relaxing the need for any local per node blind spread. No, we’re trying to protect against unpredicted load. That doesn’t mean that TE was very badly provisioned, just that it wasn’t perfect. Whatever the root cause, congestion does happen and we want to try to address it. Tony _______________________________________________ rtgwg mailing list [email protected] https://www.ietf.org/mailman/listinfo/rtgwg
