Secure Data Retrieval in Multi-Tenant Cloud Environment

Abstract

the increasing importance of cross-tenant keyword search in the fast tracks of cloud applications is aimed at perfecting their use of data. This introduces severe difficulties in terms of privacy and security, particularly in multi-tenant environments, wherein multiple owners of the data have to be considered. Most pre-existing solutions have been devised for single-tenancy systems, and they lack the ingredients for multi-tenancy. In this light, we propose a solution allowing secure keyword searching across tenants while respecting privacy concerns. Our paradigm will guarantee verifiability and accountability with respect to users' trust in search results and guarantee control over their data. Privacy will be respected by embedding the client's identity in the search, without explicit leakage of any private information. Performance is enhanced through parallel processing to speed up searches by distributing the workload across multiple partitions of the inverted index. This makes the system perform better and more scalable. We secure the encryption and decryption of data via Elliptic Curve Cryptography (ECC), and we also use SHA algorithms for deduplication of the keywords and hashing to ensure data integrity. Our system is proven to withstand a wide variety of attacks and, from our extensive experiments, we have shown that our system is capable of achieving data privacy with very high speed in a multi-tenant cloud environment. Hence, this solution fills up existing gaps and allows for secure collaborative use of the data across different tenants on the cloud.