Multi-tenant Secure Data On The Cloud:Design And Implementation

With the advent of the era of cloud computing and big data,the Internet constantly releases massive amounts of data to play a role in all walks of life.Through data integration,analysis and mining,the economic and political benefits brought by massive data are beyond people's imagination.When more and more big data appears in the cloud,problems such as the boundary problem of data use and the problem of personal privacy of data will follow one after another.The General Data Protection Regulation issued by the European Union clarifies that when a company or organization wants to collect users'private data for a certain purpose,the user's explicit consent must be obtained before that.If the company or organization wants to modify The use of the data requires the user's consent again.At the same time,when companies or organizations use data,they must not use the data to reverse the privacy of users,such as names,ID cards,or information related to genetic,economic,physical,psychological,and social identities.For the above reasons,it is difficult to directly share raw data between different organizations.Data federation provides new opportunities for data-related cooperation between different organizations by providing abstract data interfaces.With the development of cloud computing,more and more organizations store data on the cloud to gain flexibility and scalability of data processing.Existing data placement algorithms usually only consider one factor,which is either communication cost or time cost.In addition,they do not consider the characteristics of a set of jobs in the data processing process.In this article,we propose an effective solution to ensure cloud security data sharing among multiple organizations.The main achievements and innovations of this paper are as follows:(1)Aiming at the problem of cloud data storage,this paper proposes a greedy data placement algorithm based on the proposed cost model to store data in the cloud to achieve multi-objective optimization.Data storage in the cloud involves two costs:money and time.Therefore,the optimization goal of the cost model proposed in this paper consists of two parts:reducing the cost of money and the time cost of job execution at the same time.Judging from the existing research results,security strategies for data joint placement are mostly constructed based on the interdependence of data and reducing communication costs.Few strategies consider the frequency of operation.Experiments prove that the proposed algorithm effectively reduces the total cost.Based on the multi-objective cost model,we propose a greedy data placement algorithm to reduce the total cost of executing a set of operations on the cloud platform.(2)Aiming at the issue of privacy in cloud security data processing,this paper proposes a cloud scientific collaboration security data processing platform.The proposed data processing platform is used for scientific cooperation and data sharing between organizations,and completes multi-party collaboration while protecting data privacy.The platform provides users with data processing services through an identity-based proxy re-encryption mechanism.Tenants/users can upload their data to the platform and execute custom programs on Baidu Cloud.In addition,tenants/users can use data from other organizations for their own data processing work as long as they obtain the permission of the data owner.