Dimensioning network links: a new look at equivalent bandwidth

Abstract

One of the tasks of network management is to dimension the capacity of access and backbone links. Rules of thumb can be used, but they lack rigor and precision, as they fail to reliably predict whether the quality, as agreed on in the service level agreement, is actually provided. To make better predictions, a more sophisticated mathematical setup is needed. The major contribution of this article is that it presents such a setup; in this a pivotal role is played by a simple, yet versatile, formula that gives the minimum amount of capacity needed as a function of the average traffic rate, traffic variance (to be thought of as a measure of "burstiness"), as well as the required performance level. In order to apply the dimensioning formula, accurate estimates of the average traffic rate and traffic variance are needed. As opposed to the average rate, the traffic variance is rather hard to estimate; this is because measurements on small timescales are needed. We present an easily implementable remedy for this problem, in which the traffic variance is inferred from occupancy statistics of the buffer within the switch or router. To validate the resulting dimensioning procedure, we collected hundreds of traces at multiple (representative) locations, estimated for each of the traces the average traffic rate and (using the approach described above) traffic variance, and inserted these in the dimensioning formula. It turns out that the capacity estimate obtained by the procedure, is usually just a few percent off from the (empirically determined) minimally required value.

On managed services lanes and their use in home networks

“…Traffic rates higher than 40 kbit/s, up to 92 Mbit/s, have also been observed, but with a relative frequency lower than 0.002. Despite the large statistical noise, the home-network traffic-rate distribution differs significantly from that of the Internet, which generally follows a Gaussian function [13]. Our results are better approximated by a Generalized Pareto distribution, with a root mean square error of 0.0005.…”

Towards Bandwidth Estimation Using Flow-Level Measurements

“…This is especially true when flow timeouts are small. In future work we are going to embed our approach into bandwidth estimation and link dimensioning methods, such as [4,5], and assess whether applying estimated variance from flows into dimensioning formulas will still lead to good results on link dimensioning. In addition, we want to identify whether traffic distribution inside flows changes accordingly to different properties, like transport protocol and flow duration.…”

“…[4,5], and the 2011 traces were collected for these experiments. Traces from 2002 have been captured on a 300 Mbps Ethernet link that connects a residential network to the core network of a university.…”

“…For example, the work in [4,5] is based on the assumption that the traffic rate in fast networks is Normal distributed. Gaussianity is also exploited in [6].…”

Estimating Bandwidth Requirements Using Flow-Level Measurements

“…applicable not only for backbone traffic, is proposed in [8] and [9]. One of the main contributions of this work is the dimensioning formula (1).…”