NORMA eResearch @NCI Library

An Exogenous Factor Aware Resource Prediction Model for Auto-Scaling in Cloud

Shah, Priyesh (2020) An Exogenous Factor Aware Resource Prediction Model for Auto-Scaling in Cloud. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
Preview
PDF (Master of Science)
Download (1MB) | Preview
[thumbnail of Configuration manual]
Preview
PDF (Configuration manual)
Download (1MB) | Preview

Abstract

Auto-scaling is required to handle fluctuating demand in the cloud environment. It helps to add or remove resources based on pre-defined policies without human intervention. Proactive auto-scaling solutions are better than reactive ones to handle fluctuating demand but they mostly use system information and exogenous factors influencing user demand are not considered. In this paper, an exogenous factor aware resource prediction model is presented that works in the Analysis phase of the MAPE control loop. It takes into account exogenous factors to predict the user demand and its corresponding resource requirement. Historical weather and taxi booking data and weather forecast data are used for the analysis of this experiment. Taxi bookings for historical dates having weather conditions i.e. rain similar to the forecast are analyzed to predict the user demand and thereby calculate the resources required. Results show that rain has a notable impact on taxi booking as the number of taxis booked on days having rain is 49% to 61% more and the resources required are 25% to 50% more. Result validation shows that actual vs predicted taxi booking deviation is -12% to 17.5% and the actual vs predicted resource deviation is -17% to 17% which is not significant. The user demand and the resources required can be used to plan an auto-scaling operation for the upcoming days.

Item Type: Thesis (Masters)
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Cloud computing
Divisions: School of Computing > Master of Science in Cloud Computing
Depositing User: Dan English
Date Deposited: 29 Jan 2021 11:18
Last Modified: 29 Jan 2021 11:18
URI: https://norma.ncirl.ie/id/eprint/4550

Actions (login required)

View Item View Item