Improving Availability Applying Intelligent Replication in Federated Cloud Storage Based on Log Analysis

This study is focusing on improving the availability of federated storage services in order to provide better quality-of-service (QoS) to the customer with the minimum use of resources. One of the most efficient solutions to get the best experience in the cloud is to combine the services offered. In order for this to happen, there exist different approaches for selecting the best subset of services to reach the optimal performance. However, those works focus on one time selection processes, despite of customer's requirements are continuously changing and demanding adaptable storage service. In this research, I propose a method to improve storage availability through log sentiment analysis and intelligent replication. This methodology is based on the merging of two types of log analysis and the measurement of availability and performance metrics in order to select the best subset of services in cloud storage service federation.