A secure framework for managing data in cloud storage using rapid asymmetric maximum based dynamic size chunking and fuzzy logic for deduplication

Cloud storage is the ideal solution for outsourcing big data since the cloud can store a large amount of data. Cloud storage, on the other hand, raises additional problems about data duplication, fine-grained access control, and privacy, all of these factors are crucial for cloud large data storage. Data duplication approaches based on encrypted data schemes now available do not provide for fine-grained access control. This paper proposes a secure framework for managing data using rapid asymmetric maximum based dynamic size chunking and fuzzy logic for deduplication. Chunking, fingerprinting, hashing, and writing are the four main process of the proposed method. Initially, chunking is done to split the files into chunks. Rapid Asymmetric Maximum (RAM) based Dynamic Size Chunking (DSC) is used in the proposed method. These chunked files are then fingerprinted using hashing process for ensuring data authentication. Then B-tree indexing approach is used in the proposed method in order to keep the fingerprinted in an organized state. General Type2-Fuzzy logic is using Ant Lion Optimization (ALO) is used for detecting duplicate files in the documents. In the cloud storage platform, only non-duplicate documents are safely kept. The Triple Data Encryption Standard is used to do a security study before outsourcing non-duplicate data to a third-party cloud server. The total computation time of the proposed technique is 0.4 s in the inline phase and 0.04 s in the offline phase, and the deduplication ratio is 95% in the inline phase and 90% in the offline phase. This proposed deduplication approach requires less storage, which reduces memory use and processing time.