Sequence of Queues
The simplest model of a network is a sequence of queues, corresponding to the different packet transmission stages.
If we vary the packet size
- The steps are serialised, so end to end delay is the sum of time in all the queues
- The gradient of the time against length plot is the sum of the extra costs
If we consider inter-arrival times, the result is the time for the slowest server, not the sum.