Optimal Resource Allocation and Reservation using DAR in Large Scale Applications

In the current IT industry, big data analytics and Cloud Computing are the two most basic advancements. Amazingly these two innovations are come up together to give the best outcomes for different multinational business companies. In the former case, it requires huge amount of resources such as memory or hardware to store, process and other kinds of big data analytics. The cost to store this data is greatly expanded and requires innovative algorithms to reduce this complexity and this will be easy to process less information using machine learning algorithms. Distributed applications are using cloud service providers (i.e. Amazon AWS) to host and process this data with different cost to meet service level agreements. However, the customers are interested in reliable SLAs with minimized cost to store and process their data. The data centers maintained at different locations throughout the world are giving services with different get/put latencies. Allocation of data to multiple data centers and resource reservation are the two primary issues and yet to be solved. In this work, we proposed a method to reduce the cost by meeting the SLOs with integer programming. Also, we proposed an efficient method to store the data files by the optimal selection by minimizing the cost along with resource reservation. Our experimental study shows that our technique is giving the best result by selecting the optimal selection of data center along with resource reservation and its effective utilization.