DocumentCode :
299804
Title :
The management of large flows of connectionless traffic on the basis of self-similar modeling
Author :
Norros, Ilkka
Author_Institution :
VTT Information Technol/Telecommun., USA
Volume :
1
fYear :
1995
fDate :
18-22 Jun 1995
Firstpage :
451
Abstract :
Fluctuations of aggregated connectionless traffic are modelled with the fractional Brownian motion. The applicability and interpretation of the self-similar model are studied both by comparison with Ethernet traffic data from Bellcore and conceptually, defining an ideal notion of free traffic. An insight into the three model parameters is obtained by finding their counterparts in a burst model with heavy-tailed burst length distribution. A bandwidth allocation formula based on the corresponding queueing system is discussed. Guidelines for short-term traffic prediction are found using an exact formula for the conditional expectation of a future value given the past. Directions for further study are indicated at the end of each section
Keywords :
Brownian motion; Gaussian distribution; Gaussian processes; computer network management; data communication; local area networks; prediction theory; queueing theory; telecommunication traffic; Bellcore; Ethernet traffic data; bandwidth allocation formula; burst model; conditional expectation; connectionless traffic; exact formula; fractional Brownian motion; free traffic; future value; heavy-tailed burst length distribution; model parameters; queueing system; self-similar modeling; short-term traffic prediction; traffic management; Brownian motion; Channel allocation; Ethernet networks; Fluctuations; Guidelines; Information technology; Local area networks; Technology management; Telecommunication traffic; Traffic control;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Communications, 1995. ICC '95 Seattle, 'Gateway to Globalization', 1995 IEEE International Conference on
Conference_Location :
Seattle, WA
Print_ISBN :
0-7803-2486-2
Type :
conf
DOI :
10.1109/ICC.1995.525210
Filename :
525210
Link To Document :
بازگشت