The quantitative behaviour of stochastic tandem networks is considered, in which two buffers hold fluid rather than discrete tokens at each server-node. The end-to-end performance of a simple wireless router network is then optimized using such a stochastic fluid model. The optimization minimizes both the mean and variance of the transmission delay (or 'response time'), sub ject to an upper limit on the rate of losses and finite capacity queueing and recovery buffers. The trade-off between mean and variance of response time is assessed and the optimal ratio of arrival-buffer size to recovery-buffer size is determined, which is a critical quantity, affecting both loss rate and transmission time.
Information from pubs.doc.ic.ac.uk/fluid-routers.