A Cost Effective Scalable Scheme for Dynamic Data Service in Heterogeneous Cloud Environment

Abstract

Storing data in the past required a lot of resources and management. Because of tremendous utilization of web in distributed environment has created an urgent requirement for new techniques and frameworks that intelligently handled the information into valuable data and information. The past research methodologies and their frameworks focused only on static data which leads to wastage of resources and computational when the data is dynamic and soft. This paper proposed a new architectural framework to reduce the communication overhead, substantial switching cost and avoid lock-in dependency for the customers who uses the cloud services. The proposed framework uses K-means clustering analysis to filter the data in dynamic data services such as weather report, share market and streaming data like video or audio files while retrieving from the cloud. The unique feature of this framework is to provide the services using different service providers through a single interface. This research aims in enhancing the scalable resources based on the request made by the customers. The framework uses request analyzer to analyze the content which is based on media or numeric, resources are scheduled and allocated   from the providers and deliver without communication delay to the customers. The experimental results indicates the proposed framework is better option in delivering the services without communication delay.