Abstract

service level agreement (SLA) management is one of the key issues in cloud computing. The primary goal of a service provider is to minimize the risk of service violations, as these results in penalties in terms of both money and a decrease in trustworthiness. To avoid SLA violations, the service provider needs to predict the likelihood of violation for each SLO and its measurable characteristics (QoS parameters) and take immediate action to avoid violations occurring. There are several approaches discussed in the literature to predict service violation; however, none of these explores how a change in control parameters and the freshness of data impact prediction accuracy and result in the effective management of an SLA of the cloud service provider. The contribution of this paper is two-fold. First, we analyzed the accuracy of six widely used prediction algorithms—simple exponential smoothing, simple moving average, weighted moving average, Holt–Winter double exponential smoothing, extrapolation, and the autoregressive integrated moving average—by varying their individual control parameters. Each of the approaches is compared to 10 different datasets at different time intervals between 5 min and 4 weeks. Second, we analyzed the prediction accuracy of the simple exponential smoothing method by considering the freshness of a data; i.e., how the accuracy varies in the initial time period of prediction compared to later ones. To achieve this, we divided the cloud QoS dataset into sets of input values that range from 100 to 500 intervals in sets of 1–100, 1–200, 1–300, 1–400, and 1–500. From the analysis, we observed that different prediction methods behave differently based on the control parameter and the nature of the dataset. The analysis helps service providers choose a suitable prediction method with optimal control parameters so that they can obtain accurate prediction results to manage SLA intelligently and avoid violation penalties.

Highlights

  • Cloud computing is increasingly recognized and popular among business communities due to its elastic architecture and economical, accessible, scalable and flexible nature

  • According to the IBM cloud service description published in April 2019 [3], for all cloud services except Infrastructure as a Service (IaaS), the service provider is liable for 10% of service credit if the monthly uptime is less than 99.95% for services in both high availability and non-high availability zones and 25% of service credits for services if it is less than 99.90% for high availability zone and less than 99.0% for non-high availability zones

  • AUTOREGRESSIVE INTEGRATED MOVING AVERAGE METHOD (ARIMA) The method was formulated by mathematical statisticians George and Gwilym in the 1970s [71] to use with business and economic data

Read more

Summary

INTRODUCTION

Cloud computing is increasingly recognized and popular among business communities due to its elastic architecture and economical, accessible, scalable and flexible nature. In order to manage the risk of SLA violation and to avoid violation penalties, it is vital that the service provider determine the appropriate QoS prediction method based on its prediction accuracy at different time intervals. A. THE GAPS IN THE LITERATURE Firstly, it has been observed that while the existing literature evaluates various prediction approaches including machine learning, stochastic and time series prediction, none of them discusses how the different control parameters of each approach impact the prediction accuracy of cloud QoS data. None of the previous research examines how the prediction algorithms and each individual control parameters behave on cloud QoS data with different data patterns such as horizontal, cyclic and sessional at different time intervals (from 5 min to 4 weeks) and look at the prediction algorithms from a cloud SLA management perspective. The paper assists the cloud provider in choosing the optimum prediction method for different data patterns at varying time intervals.

RELEATED STUDIES
SIMPLE EXPONENTIAL SMOOTHING METHOD
WEIGHTED MOVING AVERAGE METHOD
HOLT-WINTER DOUBLE EXPONENTIAL SMOOTHING METHOD
EXTRAPOLATION METHOD
ACCURACY BENCHMARK FOR MEASURING PREDICTION ACCURACY
ANALYSING THE SES ALGORITHM BASED ON THE FRESHNESS OF DATA
Findings
DISCUSSION
CONCLUSION AND FUTURE WORK
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call