Performance Overhead Evaluation of Multi-cloud Computing with Secret Sharing approach Based Model

Abstract

Data security is one of the most critical aspects in a cloud computing environment due to the sensitivity and importance of the information stored in the cloud, as is the trustworthiness of the cloud service provider. The risk of malicious insiders in the cloud and the failure of cloud services have received intense attention by cloud users. The aim of this work is to analyse and evaluate an existing model called Multi-clouds Databases (MCDB) which uses multi-clouds instead of single cloud service provider, such as in Amazon cloud service, and compare it with other cryptographic based model. Our MCDB model incorporated Shamir’s secret sharing approach. In addition, it adopted a triple modular redundancy (TMR) technique with sequential method to improve data trustworthiness of cloud computing system and then enhance the data security aspect. The evaluation is done through simulation using cloud computing simulator. It shows a significant improvement in performance for data storage and data retrieval compared to a cloud cryptographic based model. This improvement in performance in MCDB model is due to the computational complexity of data encryption/decryption during a query execution in the cryptographic based model.