International audienceNetwork and service operators nowadays use probes located in their networks in order to improve their knowledge on traffic evolution. The limited set of managed services like IPTV enable the use of well controlled rules for network dimensioning but this is not the case for public Internet originated traffic. Due to the growing success of services delivered by Internet players, a close attention given to customers' usage is mandatory to make accurate forecasts in order to avoid future network congestion. The present paper proposes a detailed analysis based on real Internet traffic captured on fixed (xDSL, FTTH) and mobile networks of Orange France and Telefónica operators. Internet traffic profile (traffic evolution over the time) for fixed and mobile networks is described. The paper discusses the relation between access technologies and traffic profiles. Additionally, it clarifies how both fixed and mobile residential customers access Internet services. This provides insight on the applications generating the major part of the traffic (i.e. video streaming, peer-to-peer, file downloading, etc.) and on the proportion of traffic generated by the "heavy users"

We report in this paper traffic measurements of YouTube traffic from Orange networks. We specifically analyze two weeks of measurements in early April 2012. We show that the popularity curves of YouTube files are constant in time and can be well approximated by truncated Zipf laws with a shape parameter less than one. In addition, there is a huge number of files which are viewed very rarely (only once or twice). Even if this may appear as an unfavorable situation with regard to caching in view of theoretical results on cache systems, we show that thanks to file request dynamics caching is very efficient for YouTube traffic. The main reason is that files are massively requested in bursts. Since these bursts represent a significant part of traffic, caching files even by using a rather small storage capacity can achieve high gains in terms of saved bandwidth. Bursts are moreover sufficiently intense so that popular files are not pushed out of the cache memory by those files viewed only once or twice. These observations are illustrated by performing trace driven simulations by using traffic traces captured in the Orange IP backbone network.

No abstract

No abstract

No abstract