Open AccessA Ranking Based Model for Selecting Optimum Cloud Geographical Region
Author(s) -
.. Neeraj,
Major Singh,
Damanpreet Singh
Publication year - 2019
Publication title -
international journal of innovative technology and exploring engineering
Language(s) - English
Resource type - Journals
ISSN - 2278-3075
DOI - 10.35940/ijitee.j8908.0881019
Subject(s) - cloud computing , computer science , ranking (information retrieval) , overhead (engineering) , weighting , data mining , entropy (arrow of time) , cloud testing , service (business) , selection (genetic algorithm) , operations research , distributed computing , cloud computing security , engineering , machine learning , operating system , medicine , physics , economy , quantum mechanics , economics , radiology
Cloud computing has become a dominant service computing model where the services such as software, platform, infrastructure are provided to the cloud consumers on demand basis using pay as you go model. Every cloud consumer requires high performance service in minimal cost. The performance of service in cloudis measured by the parameters availability, agility, cost, and security etc. The service performance and cost are very much dependent on the cloud geographical region (CGR) where these are deployed. The services offered by the cloud service provider are installed on multiple data centers located at different CGR. In cloud environment, the selection of a service installed on the optimum CGR within the limited time overhead is a challenging and interesting problem. In this paper, the multi criteria decision making method, PROMETHEE II and objective weighting method Shannon’s Entropy,based ranking model is proposed for solving the optimum CGR selection problem. The CGR dataset of Amazon Web Service is used for the numerical analysis. The sensitivity analysis is performed for validating the stability of the proposed model and getting the most sensitive parameter. The applicability and usefulness of the service selection process is validated through the experimental results on synthetic dataset. Results show that the service selection process is achieved withlimited time overhead and hence suitable for online selection process in cloud.