Applying Autonomic Techniques To Cloud Computing For Resource Monitoring And Prediction

Abstract

Today we are in the era of distributed system and cloud computing is the best example for the same. Resource management is an important issue in cloud computing, we need to keep track of the available resources in cloud, so that we can give services to user to fulfill their requirement. Which leads in helping to generate maximum revenue, lower powerconsumption, carbon emission and ultimately leads to green computing. So with the help of resource monitoring we can get the data about how, when, in what amount of resources for a particular cloud and the user. With effective resource monitoring, by minimizing some monitoring units we can reduce the cost of monitoring in terms of computation and power consumption. Our main objective is to reduce the monitoring technique, so that the amount of computation and power consumption can be saved which will lead to smart and green computing. In this project we had proposed an algorithm for reducing the monitoring overhead for cloud computing. Resource prediction can make resource management much more easier. Prediction techniques like machine learning or neural networks can be very healpfull for predicting our cloud ressources. In this project Long Short Term Memory (LSTM) is being applied for the resource prediction in cloud computing. So with the help of resource monitoring and predition in cloud we can manage the cloud resources very effectively.