In lecture, this slide showed a single packet being sent from the source to the destination. The packet took 4 units of time to copy from one node to the next, with the link between the source and target node being marked as a "Busy Link" during the transfer. We complete the entire transmission in 12 time steps, 3 node transfers times 4 time steps per node transfer.
I suspect that this might give poor performance if the cost to travel through the links is different since links will sit idle with the packet unable to make progress if we are waiting on a long latency link.
In lecture, this slide showed a single packet being sent from the source to the destination. The packet took 4 units of time to copy from one node to the next, with the link between the source and target node being marked as a "Busy Link" during the transfer. We complete the entire transmission in 12 time steps, 3 node transfers times 4 time steps per node transfer.
I suspect that this might give poor performance if the cost to travel through the links is different since links will sit idle with the packet unable to make progress if we are waiting on a long latency link.
This comment was marked helpful 0 times.