METHOD FOR PREDICTING THE EXECUTION TIME OF A PLURALITY OF VIRTUAL MACHINES IN A COMPUTING CLOUD

The present invention aims to predict the average, the minimum and maximum execution time of a plurality of virtual machines (VMs), considering several impact factors in the execution thereof. Also, it consists of collecting impact factor data, preprocessing and grouping the collected data into diff...

Full description

Saved in:
Bibliographic Details
Main Authors DURÁN LIMÓN Héctor Alejandro, GOVINDARAJU Yatheendraprakash, MEZURA MONTES Efrén
Format Patent
LanguageEnglish
Spanish
Published 11.01.2022
Subjects
CALCULATING
COMPUTING
COUNTING
ELECTRIC DIGITAL DATA PROCESSING
PHYSICS
Online AccessGet full text

Cover

More Information
Summary:The present invention aims to predict the average, the minimum and maximum execution time of a plurality of virtual machines (VMs), considering several impact factors in the execution thereof. Also, it consists of collecting impact factor data, preprocessing and grouping the collected data into different sets, generating a regression tree for each set, pruning each regression tree, and using them to make predictions of VM execution times. La presente invención busca predecir el tiempo de ejecución promedio, mínimo y máximo de una pluralidad de máquinas virtuales (MVs); y considerar una pluralidad de factores de impacto en la ejecución de máquinas virtuales, implementando para ello un método que puede comprender las etapas de: recolectar datos de factores de impacto, preprocesar y agrupar los datos recolectados en diferentes conjuntos, generar un árbol de regresión para cada conjunto, podar cada árbol de regresión, y usar cada árbol para hacer predicciones de los tiempos de ejecución de una MV.
Bibliography:Application Number: MX20210006157