SECURE AND EFFICIENT DATA DEDUPLICATION IN JOINT CLOUD STORAGE

Abstract

The exponential growth of data in cloud storage has led to significant challenges in storage efficiency, bandwidth consumption, and management costs. Data deduplication is a proven technique that eliminates redundant copies of data, storing only a single instance while reducing storage space and upload bandwidth. However, in multi-cloud or Joint Cloud Storage environments (where multiple cloud providers collaborate to offer global services), achieving secure deduplication becomes complex due to privacy concerns, key management issues, and dynamic ownership changes. This project presents SED (Secure and Efficient Deduplication) — a novel scheme designed specifically for Joint Cloud Storage systems. It supports secure cross-user deduplication without relying on a fully trusted key server. The system employs convergent encryption combined with advanced ownership management and dynamic access control to ensure that identical data is stored only once while preserving data confidentiality. It handles dynamic operations such as user revocation, ownership updates, and efficient verification. By integrating hybrid cloud concepts (public cloud for storage and private cloud for ownership control), the proposed system resists collusion attacks, duplicate faking attacks, and side-channel leaks. Experimental evaluation shows substantial savings in storage space (up to 70–90% in redundant datasets) and reduced communication overhead compared to traditional schemes. The solution is scalable, secure, and ideal for enterprise Joint Cloud environments seeking optimized storage with strong privacy guarantees.