Kohonen Self-Organizing Maps and Expert System for Network Virtual Memory Performance Prediction

A basic reason why we monitor the performance of a system is to define the overall state of the machine ("It's Ok or "It's slow" etc). In other words to determine whether the system is experienceing a bottleneck or not. This paper investigates the application of Self Organiz...

Full description

Saved in:
Bibliographic Details
Published inSystems analysis, modelling, simulation Vol. 42; no. 7; pp. 1025 - 1043
Main Authors Elfadil, Nazar, Hani, Mohamed Khalil, Nor, Sulaiman Mohd, Hussein, Sheikh
Format Journal Article
LanguageEnglish
Published Taylor & Francis Group 01.01.2002
Subjects
Expert System
Knowledge Acquisition
Kohonen Self-organizing Maps
Machine Learning
Network Performance
Rule Extraction
Online AccessGet full text

Cover

More Information
Summary:A basic reason why we monitor the performance of a system is to define the overall state of the machine ("It's Ok or "It's slow" etc). In other words to determine whether the system is experienceing a bottleneck or not. This paper investigates the application of Self Organizing Maps (SOM) and Expert System (ES) in the network performance management area. A system is proposed which collects data from certain server and processes the data in order to get knowledge concerning the current network performance or predicts the network performance. To demonstrate the applicability of the proposed method to network performance domain problems, a case study in virtual memory performance status (vmstat) domain is presented.
ISSN:0232-9298
1029-4902
DOI:10.1080/716067205