A Reinforcement Learning Based Auto-Scaling Approach for SaaS Providers in Dynamic Cloud Environment

Yi Wei; Daniel Kudenko; Shijun Liu; Li Pan; Lei Wu; Xiangxu Meng

doi:10.1155/2019/5080647

A Reinforcement Learning Based Auto-Scaling Approach for SaaS Providers in Dynamic Cloud Environment

Yi Wei, Daniel Kudenko, Shijun Liu, Li Pan, Lei Wu, Xiangxu Meng

Computer Science

Research output: Contribution to journal › Article › peer-review

Abstract

Cloud computing is an emerging paradigm which provides a flexible and diversified trading market for Infrastructure-as-a-Service (IaaS) providers, Software-as-a-Service (SaaS) providers, and cloud-based application customers. Taking the perspective of SaaS providers, they offer various SaaS services using rental cloud resources supplied by IaaS providers to their end users. In order to maximize their utility, the best behavioural strategy is to reduce renting expenses as much as possible while providing sufficient processing capacity to meet customer demands. In reality, public IaaS providers such as Amazon offer different types of virtual machine (VM) instances with different pricing models. Moreover, service requests from customers always change as time goes by. In such heterogeneous and changing environments, how to realize application auto-scaling becomes increasingly significant for SaaS providers. In this paper, we first formulate this problem and then propose a Q-learning based self-adaptive renting plan generation approach to help SaaS providers make efficient IaaS facilities adjustment decisions dynamically. Through a series of experiments and simulation, we evaluate the auto-scaling approach under different market conditions and compare it with two other resource allocation strategies. Experimental results show that our approach could automatically generate optimal renting policies for the SaaS provider in the long run.

Original language	English
Article number	5080647
Number of pages	11
Journal	Mathematical Problems in Engineering
Volume	2019
DOIs	https://doi.org/10.1155/2019/5080647
Publication status	Published - 3 Feb 2019

Bibliographical note

Access to Document

10.1155/2019/5080647

5080647 (1)
© 2019 Yi Wei et al.
Final published version, 3.26 MBLicence: CC BY

Cite this

@article{55bbee027615457d8b6467d43daaad01,

title = "A Reinforcement Learning Based Auto-Scaling Approach for SaaS Providers in Dynamic Cloud Environment",

abstract = "Cloud computing is an emerging paradigm which provides a flexible and diversified trading market for Infrastructure-as-a-Service (IaaS) providers, Software-as-a-Service (SaaS) providers, and cloud-based application customers. Taking the perspective of SaaS providers, they offer various SaaS services using rental cloud resources supplied by IaaS providers to their end users. In order to maximize their utility, the best behavioural strategy is to reduce renting expenses as much as possible while providing sufficient processing capacity to meet customer demands. In reality, public IaaS providers such as Amazon offer different types of virtual machine (VM) instances with different pricing models. Moreover, service requests from customers always change as time goes by. In such heterogeneous and changing environments, how to realize application auto-scaling becomes increasingly significant for SaaS providers. In this paper, we first formulate this problem and then propose a Q-learning based self-adaptive renting plan generation approach to help SaaS providers make efficient IaaS facilities adjustment decisions dynamically. Through a series of experiments and simulation, we evaluate the auto-scaling approach under different market conditions and compare it with two other resource allocation strategies. Experimental results show that our approach could automatically generate optimal renting policies for the SaaS provider in the long run.",

author = "Yi Wei and Daniel Kudenko and Shijun Liu and Li Pan and Lei Wu and Xiangxu Meng",

note = "{\textcopyright} 2019 Yi Wei et al.",

year = "2019",

month = feb,

day = "3",

doi = "10.1155/2019/5080647",

language = "English",

volume = "2019",

journal = "Mathematical Problems in Engineering",

issn = "1563-5147",

publisher = "Hindawi Publishing Corporation",

}

TY - JOUR

T1 - A Reinforcement Learning Based Auto-Scaling Approach for SaaS Providers in Dynamic Cloud Environment

AU - Wei, Yi

AU - Kudenko, Daniel

AU - Liu, Shijun

AU - Pan, Li

AU - Wu, Lei

AU - Meng, Xiangxu

PY - 2019/2/3

Y1 - 2019/2/3

N2 - Cloud computing is an emerging paradigm which provides a flexible and diversified trading market for Infrastructure-as-a-Service (IaaS) providers, Software-as-a-Service (SaaS) providers, and cloud-based application customers. Taking the perspective of SaaS providers, they offer various SaaS services using rental cloud resources supplied by IaaS providers to their end users. In order to maximize their utility, the best behavioural strategy is to reduce renting expenses as much as possible while providing sufficient processing capacity to meet customer demands. In reality, public IaaS providers such as Amazon offer different types of virtual machine (VM) instances with different pricing models. Moreover, service requests from customers always change as time goes by. In such heterogeneous and changing environments, how to realize application auto-scaling becomes increasingly significant for SaaS providers. In this paper, we first formulate this problem and then propose a Q-learning based self-adaptive renting plan generation approach to help SaaS providers make efficient IaaS facilities adjustment decisions dynamically. Through a series of experiments and simulation, we evaluate the auto-scaling approach under different market conditions and compare it with two other resource allocation strategies. Experimental results show that our approach could automatically generate optimal renting policies for the SaaS provider in the long run.

AB - Cloud computing is an emerging paradigm which provides a flexible and diversified trading market for Infrastructure-as-a-Service (IaaS) providers, Software-as-a-Service (SaaS) providers, and cloud-based application customers. Taking the perspective of SaaS providers, they offer various SaaS services using rental cloud resources supplied by IaaS providers to their end users. In order to maximize their utility, the best behavioural strategy is to reduce renting expenses as much as possible while providing sufficient processing capacity to meet customer demands. In reality, public IaaS providers such as Amazon offer different types of virtual machine (VM) instances with different pricing models. Moreover, service requests from customers always change as time goes by. In such heterogeneous and changing environments, how to realize application auto-scaling becomes increasingly significant for SaaS providers. In this paper, we first formulate this problem and then propose a Q-learning based self-adaptive renting plan generation approach to help SaaS providers make efficient IaaS facilities adjustment decisions dynamically. Through a series of experiments and simulation, we evaluate the auto-scaling approach under different market conditions and compare it with two other resource allocation strategies. Experimental results show that our approach could automatically generate optimal renting policies for the SaaS provider in the long run.

U2 - 10.1155/2019/5080647

DO - 10.1155/2019/5080647

M3 - Article

VL - 2019

JO - Mathematical Problems in Engineering

JF - Mathematical Problems in Engineering

SN - 1563-5147

M1 - 5080647

ER -