Prediction of SLA Violation in Cloud Resource Allocation using Machine Learning Based Back Propagation Neural Network (BPNN)
Karthik Kambhampati1, A. Srinagesh2

1Karthik Kambhampati, Research Scholar, Department of Computer Science and Engineering, Acharya Nagarjuna University, Guntur (Andhra Pradesh), India.
2Dr.A. Srinagesh, Associate Professor, Department of Computer Science and Engineering, RVR&JC College of Engineering, Guntur (Andhra Pradesh), India.
Manuscript received on 03 June 2019 | Revised Manuscript received on 07 June 2019 | Manuscript published on 30 June 2019 | PP: 2109-2114 | Volume-8 Issue-8, June 2019 | Retrieval Number: H7536068819/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: Cloud computing is defined as a pay-per-use model which would offer the user required services. Cloud computing services can be identified as three models based on the type of resources and the services they are provisioning. Cloud Service models that are more popular in cloud computing environment are, “SaaS, IaaS, and PaaS”. Among these cloud service models, SaaS model provides flexible and reliable services to the cloud users based on their requirement. SaaS rents the required resources from the IaaS cloud Service providers to offer the services to the cloud users in a reliable manner. However, this rental scheme would increase the administration and maintenance costs. And also, resources that are rented from the external cloud resource providers might degrade the service quality in hosting services due to lack of SLA agreement. Several research works have been conducted previously for performing a better admission control. Some of the research works are discussed here for better understanding of the merits and demerits of existing research efforts. In this paper we are focusing on SLA violation prediction by both user as well as CSP. For to predict SLA violation we use machine learning based back propagation network model. The experimental results shows that BPNN performed well when compared to the statically models.
Index Terms: Cloud, SLA, BPNN, Cloud Service Provider, Resource Allocation.

Scope of the Article: Machine Learning