Premium
The calibration and validation of a hybrid simulation/regression model of a batch computer system
Author(s) -
Gomaa H.
Publication year - 1978
Publication title -
software: practice and experience
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.437
H-Index - 70
eISSN - 1097-024X
pISSN - 0038-0644
DOI - 10.1002/spe.4380080103
Subject(s) - calibration , workload , computer science , trace (psycholinguistics) , regression analysis , model validation , simulation , statistics , machine learning , mathematics , linguistics , philosophy , data science , operating system
Abstract Calibration and validation are two stages in the development of a performance model of a computer system during which the degree of confidence in the model is established. This paper describes a case study of the calibration and validation of a hybrid simulation/regression Load Adjusting Model of a batch computer system. During calibration, a given workload trace was applied to the Load Adjusting Model and the parameters of the model were adjusted using an iterative tuning procedure. During validation, the parameters of the model, established during calibration, were kept constant and the model was tested using different workload traces.