PROJECT TITLE :

Impact of Packet Sampling on Link Dimensioning

ABSTRACT:

Link dimensioning is used by network operators to properly provision the capability of their network links. Proposed ways for link dimensioning usually require statistics, such as traffic variance, that require to be calculated from packet-level measurements. In observe, due to increasing traffic volume, operators deploy packet sampling techniques aiming to scale back the burden of traffic monitoring, however very little is thought concerning how link dimensioning is affected by such measurements. During this paper, we have a tendency to make use of a previously proposed and validated dimensioning formula that requires traffic variance to estimate required link capacity. We assess the impact of three packet sampling techniques on link dimensioning, particularly, Bernoulli, n-in-N and sFlow sampling. To account for the additional variance introduced by the sampling algorithms, we tend to propose approaches to higher estimate traffic variance from sampled data in line with the used technique. Results show that, depending on sampling rate and link load, packet sampling does not negatively impact on link dimensioning accuracy even at terribly short timescales like ten ms. Moreover, we conjointly show that the loss of inter-arrival time of sampled packets due to the exporting process in sFlow will not harm the estimations, on condition that an appropriate sampling rate is used. Our study is validated employing a massive dataset consisting of traffic packet traces captured at many locations around the world.