Improving Data Transfer Performance to Unlock Potential of Cloud Service Providers

Abstract

Efficient data transfer among CSPs is essential for the performance of cloud computing's basic operations, such as migration and disaster recovery. In this research, a novel approach to enhance data transmission performance using DTNs is presented. Using DTNs wisely, the proposed method enhances the speed and reliability of sending huge volumes of data across several CSPs. Coordinating local and remote copy operations is also a part of the process as well as a mechanism for DTN-to-DTN transfers. Optimized network settings with NFS connections maximized data transfer speed. In order to optimize the network, we carried out a battery of experiments to determine the best setting by varying network buffer size, synchronous and asynchronous modes, among others. With asynchronous modes and the best possible NFS settings, results showed a dramatic improvement in data transmission speed. The method is scalable and reliable when looking to address the challenge of inter-cloud improvements with regard to data transfer within a multi-cloud environment.