Web server workload forecasting in fuzzy environment - linguistic approach

Author

Cheong, Chin Wen ; Way, Chong Chee ; Ramachandran, V.

Author_Institution

Fac. of Inf. Technol., Multimedia Univ., Selangor, Malaysia

Volume

1

fYear

2000

fDate

2000

Firstpage

361

Abstract

Web server workload forecasting is one of the essential considerations in Web server management and network upgrading. Due to variability of server workload distribution originated from unpredictable users´ surfing behavior, the measurement of Web server performance metrics are characterized and modeled in fuzzy manner. A fuzzy inference system is formed using four Web server performance metrics and a server utilization index are derived to determine the servers´ utilization states for every time period. A fuzzy Markov model is proposed to illustrate the state transitions of server resource utilization based on expert linguistic evaluation of stationary transition probability. A steady state algorithm is applied to explore the convergence of server resource utilization after n transition periods

Keywords

Internet; Markov processes; computer network management; fuzzy logic; knowledge based systems; network servers; performance evaluation; Markov chain; Web server managemen; Web server workload forecasting; fuzzy Markov model; fuzzy environment; fuzzy inference system; fuzzy logic; linguistic approach; network upgrading; performance metrics; server performance metric; server resource utilization; server utilization index; server workload distribution; state transitions; stationary transition probability; steady state algorithm; user surfing behavior; Artificial intelligence; Artificial neural networks; Delay; Economic forecasting; Fuzzy logic; Information technology; Measurement; Resource management; Web server; World Wide Web;

fLanguage

English

Publisher

ieee

Conference_Titel

TENCON 2000. Proceedings

Conference_Location

Kuala Lumpur

Print_ISBN

0-7803-6355-8

Type

conf

DOI

10.1109/TENCON.2000.893690

Filename

893690