Caracterization of traffic in a local area network
No Thumbnail Available
Date
Advisors
Evaluators
Journal Title
Journal ISSN
Volume Title
Publisher
Universidad Industrial de Santander
Abstract
Description
Este trabajo de investigación se orienta al análisis de un segmento de una LAN, con el fn de caracterizar el tráfco entrante y saliente a servidores para posteriormente contrastarlo mediante el uso de un simulador de tráfco. Para lograrlo, se registran datagramas de diferentes servidores correspondientes a protocolos comúnmente utilizados como el SMTP1, HTTP2, entre otros. Con dichos datagramas se determina el grado de auto-similitud y las características probabilísticas que presenta cada servidor, utilizando diferentes pruebas para cada caso. Una vez caracterizado el tráfco de los servidores, se utiliza una herramienta de simulación para conseguir un ambiente de tráfco similar al real.
This research work presents an analysis of the traffc on a LAN segment heading to model its behavior and the generation of similar results by using a traffc simulation tool. To achieve this, datagrams are collected from servers supporting commonly used protocols such as SMTP1 and HTTP2. The datagrams are then used to determine the degree of self-similarity of the network segment and its probabilistic characteristics. Different tests are used for each case. Once the network segment model is obtained, a simulation tool is used to get a similar traffc environment to the real network traffc.
This research work presents an analysis of the traffc on a LAN segment heading to model its behavior and the generation of similar results by using a traffc simulation tool. To achieve this, datagrams are collected from servers supporting commonly used protocols such as SMTP1 and HTTP2. The datagrams are then used to determine the degree of self-similarity of the network segment and its probabilistic characteristics. Different tests are used for each case. Once the network segment model is obtained, a simulation tool is used to get a similar traffc environment to the real network traffc.
Keywords
traffic, self-similarity, long range dependence, Hurst parameter, Whittle estimator, periodogram, Autosimilaridad, dependencia a largo plazo, parámetro de Hurst, estimador de Whittle, periodograma