CERN Accelerating science

If you experience any problem watching the video, click the download button below
Download Embed
Internal Note
Report number CERN-IT-Note-2017-001
Title Simulating network throughput by correlating perfSONAR measurements with link utilisation
Author(s) Babik, Marian (CERN) ; Hendrik, Borras (Ruprecht-Karls-Universitaet Heidelberg (DE))
Corporate author(s) CERN. Geneva. IT Department
Publication 2017
Imprint 14 Feb 2017
Number of pages 22
Subject category Computing and Computers
Keywords perfSONAR ; throughput ; simulation ; machine learning ; network ; analytics
Abstract In this report the modeling of the network performance through delay and packet loss measurements is explored. It is shown that correlations exists and different models are proposed to exploit them. In addition, the developed models are then implemented into the existing network measurement platform and made available to general use.
Submitted by [email protected]

 


 Datensatz erzeugt am 2017-02-14, letzte Änderung am 2017-02-14


Access to fulltext:
Volltext herunterladen
DOCX.PDF