Cloud data. Data in cloud storage or online storage.

Cloud data. For example, in cloud storage or online storage through Dropbox, iCloud, Securesafe or Speicherbox, data is stored centrally on the Internet. This system has several advantages, but also risks. In the end, you give your data to third parties! The geographic area of the cloud provider is an important selection criterion. When your data is stored abroad, it is subject to another data protection law.

Cloud data storage. Choose the right cloud provider. Foreign cloud providers, and in particular non-European cloud providers, sometimes have data protection weaknesses. Choose secure access. When you have the opportunity, use two-factor authentication. Follow our 6 rules to create a strong password.

Cloud data encryption. Transmit your data only in encrypted form. Select a service where the data transfer is encrypted. For example, in your browser, you can confirm this by checking that the address bar starts with https: // and the lock symbol is displayed. Keep your data encrypted only. In the general case, it is impossible to verify that your data is indeed encrypted by the cloud provider. Therefore, it is best to take care of the encryption and decryption of your sensitive data. Protect your data on a local drive. As a rule, it is also impossible to verify that your data has been correctly copied by the cloud provider. It is better to make regular local backups of data stored in the cloud.

What is cloud storage? Cloud storage or online storage is storage available only online. Data stored in the cloud (documents, photos, movies, passwords, etc.) can be accessed (simultaneously) from multiple devices regardless of where you are. This mobile access allows you, for example, to view, share and edit photos even on the go. Finally, this solution allows several people in different places to work on the same document.

With the development of the Internet, computing has been based mainly on communication between servers, user stations, networks and data centers. In the early 2000s, there were two trends in application delivery and infrastructure virtualization. The convergence of these two trends has led to a unifying concept of cloud computing. Thus, data storage is a central element of the problem of deploying processes and resources in the cloud. Data security – whether it is simple outsourcing of storage for backup purposes, the use of hosted software services or virtualization with a third-party provider of company IT infrastructure. crucial.

This security is divided into three areas: data availability, integrity and confidentiality. The context of our work concerns storage virtualization for cloud computing. This work is part of the Secured Virtual Cloud (SVC) project, funded by the National Future Investment Digital Society Fund. They led to the development of middleware for storage virtualization called CloViS (Cloud Virtualized Storage), which is entering the evaluation phase based on the SATT Toulouse-Tech-Transfer. CloViS is a middleware for data management developed in the IRIT lab that allows you to virtualize heterogeneous and distributed storage resources that are accessible in a uniform and transparent way.

The feature of CloViS is to meet the needs of users and system availability thanks to the quality of service defined for virtual volumes. Our contribution to this area is related to data distribution methods to increase their availability and reliability of input-output operations in CloViS. Indeed, in the face of a data explosion, the use of replication cannot be a long-term solution. The use of corrective codes or threshold value schemes is then presented as a valid alternative to control storage volumes.

However, at present no protocol has been adapted to these new dissemination methods to ensure data consistency. For this purpose, we offer data consistency protocols adapted to these various data distribution methods. We then analyze these protocols to highlight their advantages and disadvantages. Indeed, the choice of data distribution method and the corresponding data consistency protocol is based on performance criteria, including read and write accessibility, use of system resources (such as used storage), or the average number of messages exchanged during read and write operations.