AN ENERGY EFFICIENT AND SELF ADAPTIVE RESOURCE ALLOCATION FRAMEWORK USING MODIFIED CLONAL SELECTION ALGORITHM FOR CLOUD BASED SOFTWARE SERVICES

Cloud computing is a prominent model for computation which lays greater impact on IT industries and the way in which software applications are developed and deployed. The workload in cloud computing changes dynamically and thereby introduces many challenges in on-demand resource provisioning and allocation. Resource allocation techniques are not widely available. Those that exist also have high energy consumption as well as the ineffectiveness of allocation. Frequent virtual machine switches in Virtual Machine allocation leads to a tradeoff between QoS and consumption of energy. Hence there is a need to find a solution which provides a quality of service and low energy consumption for use in cloud services. The proposed work suggests self-adaptive framework for allocation of resources composed of feedback loops to meet the QoS requirements as well as less energy consumption in cloud computing services. The self-adaptive resource allocation framework comprises of three stages namely QoS prediction model, improved cuckoo search algorithm (ICSA) - based runtime decision algorithm and Energy Efficient Model (EEM). In the First stage, QoS prediction model works over historical data of a system that aids in improving the accuracy rate of QoS prediction. Second an Energy Efficient Model based on Modified Clonal Selection Algorithm (MCSA) has been suggested for reducing the energy consumption. Third stage is a runtime decision-making algorithm which works according to the ICSA helps to find proper operation for allocating resource in an online real context. Experiments are conducted and the results convey that the proposed work could decrease the number of times hosts are switched on/off. This technique when compared to existing methods, helps to save power, provides cost effective and improves the QoS for cloud computing environments.