Probabilistic Deterministic Load Balancing Analysis for File Sharing in Cloud Environment

Cloud data center consists of a network of heterogeneous multiple servers. Cloud-based applications depend heavily on load balancing and optimization of file sharing among its user base. Optimal load balancing methods become more demanding, with increased cloud service provisioning. However, little research has been undertaken to handle the complex load distribution (indeterminacy) among the cloud servers in terms of data utility and delivery. In this paper, we introduce a technique to analyze indeterminacy load in the cloud data server with the evolution of random values over time. Probabilistic Deterministic Load Balancing (PDLB) method is introduced for efficiently share the files with minimal response time in cloud infrastructure. A Load balancing architecture based on Probabilistic deterministic Process is designed to minimize cloud user response time for all users in Cloud infrastructure. Based on this method, a Probabilistic Deterministic based Load Balancing algorithm is designed. Next, load balancing is performed according to the aggregated load of each cloud server based on the probability deterministic output. Performance evaluation is conducted with Amazon EC2 for evaluating the performance of file sharing. The results of trace-driven experiments on the real-world Amazon EC2 demonstrate higher efficiency, trustworthiness, minimizing the response time and file sharing overhead compared with other methods.