Premium
9.1.7 Prediction of SAP based ERP Systems Capacity Using Stochastic Processes and Spectrum Analysis
Author(s) -
Zuniga Roberto B.
Publication year - 2001
Publication title -
incose international symposium
Language(s) - English
Resource type - Journals
ISSN - 2334-5837
DOI - 10.1002/j.2334-5837.2001.tb02434.x
Subject(s) - computer science , set (abstract data type) , signal (programming language) , digital signal processing , database transaction , transformation (genetics) , variable (mathematics) , signal processing , stochastic process , event (particle physics) , algorithm , real time computing , mathematics , computer hardware , database , statistics , mathematical analysis , biochemistry , chemistry , physics , quantum mechanics , gene , programming language
This paper presents an innovative approach to predict system resources consumption within an ERP production environment running SAP R/3 applications. It combines ‘Discrete Event’ modeling technologies with Digital Signal Processing (DSP) methodologies for analysis of stochastic (random variable described) phenomena. The approach consists in considering the samples collected by system monitoring tools as discrete stochastic signals representing system resources behavior. These signals are filtered and processed by Time – Frequency transformation algorithms to extract a set of sinusoidal mathematical sequences that characterize the periodicity of each signal. These set of discrete frequency components uniquely identifies system resources (CPU, Disk IO, etc) behavior as effected by the transaction been executed.