Scalability and Efficiency of Homomorphic Encryption in Cloud Computing: Overcoming Challenges in Secure Data Processing

Abstract

Homomorphic Encryption (HE) is another privacy-preserving enabler for cloud computing because it enables one to execute operations on encrypted data without actually decrypting it. In this work, various HE schemes such as FHE, FHE with Bootstrapping, PHE, and PHE with Parallelism are experimented in terms of scalability and performance under the Cloud Computing emulation environment. Wherever possible, we compare the execution parameters such as time, space should also be evaluated in addition to noise amplification for each scheme in varying data scenarios and task sizes. Its findings state that while FHE provides a high level of security, it compromises significantly in terms of computation and memory usage, particularly when dealing with a large dataset. Bootstrapping increases these inefficiencies while PHE scales better and requires less resources especially when parallel computing is utilized. The outcomes are analysed with respect to prior literature and the presentation of these approaches to improve HE with real-world cloud application using optimization methods like noise management, parallelization, and mixed encryption techniques. In conclusion, the paper clearly points out that when it comes to achieving optimal system performance and security, it is important to combine HE with distributed architectures and other efficient cryptographic approaches, so that it can effectively serve as the foundation for secure cloud-supported data processing solutions.