Traffic Management for the Available Bit Rate (ABR) Service in ...

We argued that the queue length is not a good indicator of load when the control is
rate-based.
Figure 5.5: Congestion Detection Metric: Queue Length or Input Rate ?
As an example, consider tworatecontrolled queues as shown in gure 5.5. Suppose
the rst queue is only 100 cells long while the other is 1000 cells long. Without further
information it is not possible to say which queue is overloaded. For example, if the
rst queue is growing at the rate of 1000 cells per second, it is overloaded while the
second queue may be decreasing at a rate of 1000 cells per second and may actually be
underloaded. Further, if the rst queue can be processed at 622 Mbps, the queueing
delayismuch smaller than that of a 100 cell queue processed at 1.54 Mbps. This factor
becomes important because the capacity available to ABR can be quite variable.
Another important reason for the choice of the input rate metric has to do with
rate and window controls. For a detailed discussion of rate versus window, see Jain
(1990) [48]. In particular, a window controls the queue length, while the rate controls
the queue growth rate. Given a particular window size, the maximum queue length
can be guaranteed to be below the window. Given an input rate to a queue, the queue
growth rate can be guaranteed below the input rate but there is nothing that can be
109