Cost Effective and Scalable Service Selection in Heterogeneous Cloud

Abstract

Cloud computing is a powerful computing paradigm to deliver services over newlinethe internet. It is a model that facilitates on demand network access to a shared pool of newlineconfigurable computing resources that can be rapidly provisioned and released with newlineminimal management effort or service provider interaction. Scalability is one the main newlinefeatures of cloud computing that supports long term strategies and business needs. newlineAlthough, cloud computing is equipped with high performance components, it is newlinenoted that lack of resource allocation and efficient task scheduling. Resource newlineallocation is a very difficult task for cloud based services due to the nature of requests newlinelike dynamic, unpredictable and uncertainty etc. newlineThis research work designed different architectures to enhance scalability newlinefeatures for dynamic data, media sharing and social networking service in newlineheterogeneous cloud environment. The developed frameworks are as follows: newline1. Cost Effective Scalable Framework for Dynamic Data (CESFDD) service newline2. Cost Effective Scalable Scheme for Media Streaming (CESSMS) service newline3. Cost Effective Scalable Scheme for Auto scaling based on Dynamic Threshold newline(CESSADT) newlineFirst phase deals with the development of architecture, CESFDD in newlineheterogeneous cloud environment in order to reduce communication overhead, newlinesubstantial switching and total cost to cloud users. The developed CESFDD newlineframework allocates resources based on events raised by the users which guarantee newlineefficient communication. In CESFDD, request analyzer accepts service request from newlinethe user add to check the type of request. Resource allocator assigns the number of newlineresources required to complete the request. Resource coordinator allocates time and newlinesubmits the request to both proxy server and third party service providers. If the newlineservice is found in the physical proxy server then response is passed to the user else newlinethe user obtains service from the third party service providers. This process reduces newlineresponse time and communication delay. It also uses d