Secure Distributed Data using Multi-Cloud

Abstract

Abstract: Abstract‫ــــ‬Implementing cloud computing provides many paths for web-based service. But, data security and privacy requirement become an important problem that limits several cloud applications. One of the key security and privacy concerns is the fact that cloud service suppliers have access to data. This concern greatly reduces the usability of cloud computing in many areas, such as financial business and government agencies. This paper focuses on this important issue and suggests a new approach, so cloud providers cannot directly access data. The proposed approach is divided into two sides: upload side and download side. In upload side, there is three stages, at the first stage; the transmitted file is splitted and then encrypted in order to achieve the data security requirement. At the second stage, the splitted data are integrity checked by MD5 algorithm, in order to achieve integrity requirement. At the third stage, the checked splitted data are stored separately in three -clouds, in order to achieve distribution requirement. In download side, also there is three stages. At the first stage, the data is retrieved from the three-clouds. At the second stage, data integrity is performed using MD5. At the third stage, data decryption and merging are done. The proposed approach is successfully implemented on (25 KB) image. The proposed model is successfully implemented in uploading side dependent on shares3 because provide high security with total time of (8.144 sec), and in downloading side with total side of (9.42).