An efficient architecture and algorithm for server provisioning in Cloud computing using clustering approach

摘要

Cloud computing is a method for distributed system where software and hardware resources will be distributed through the networked Cloud and all services fetches together on the source of user requirements as they need. In cloud computing Load Balancing is always a big challenge among network researchers. This is because if network load is equally divided onto the entire network it gives faster connectivity on application level as well on hardware level. Fog Computing represents a new model that extends Cloud computing and services to the border of the network. Fog is comparable to cloud it provides data and application facility to end-users. There is one difference lies in the information that it is located more close to end users to process and give reply to the clients/users in less time. To make Cloud computing more responsive and robust load balancing must be implemented on cloud servers It is a useful technique to reduce time complexity, efficiency of server and specially job response times, under large and unbalanced workload. Primarily the centralized model was used for the load balancing in scattered environment. Though, there is still a limitation that it consumes more time to finds the nearest server for transferring the load, when the server is overloaded. This paper proposes an efficient architecture and algorithm for server provisioning in Cloud computing environment using clustering approach.