Design and Implementation of a Secure Multi-Cloud Data Storage System Using Hybrid Cryptography and Threshold Secret Sharing

Abstract

Cloud computing has fundamentally transformed how individuals, enterprises, and government agencies store, manage, and access data. While cloud storage services provide scalable, flexible, and cost-effective solutions, traditional single-cloud storage models present inherent vulnerabilities including single points of failure, vendor lock-in, insider threats, service outages, and privacy breaches. This research presents the design and implementation of a Secure Multi-Cloud Data Storage System (SMCDSS) that leverages hybrid cryptography, data fragmentation, and threshold secret sharing techniques to enhance data confidentiality, integrity, availability, and fault tolerance across multiple independent cloud service providers. The system integrates AES symmetric encryption for data protection, RSA/ECC asymmetric encryption for secure key exchange, and Shamir's threshold secret sharing scheme for distributed key management. Data integrity is verified using SHA-256 hashing algorithms. The proposed architecture distributes encrypted data fragments across multiple cloud providers, ensuring that no single provider possesses complete information. Performance evaluation demonstrates significant improvements in security resilience, fault tolerance, and data availability compared to traditional single-cloud storage models. The system is implemented as a prototype web-based application with a user-friendly interface for secure file upload, retrieval, and monitoring. The findings indicate that multi-cloud architectures combined with robust cryptographic mechanisms offer a powerful solution for secure distributed data storage, providing valuable support for organizations seeking to protect sensitive information in cloud environments. Future research will focus on integration with real cloud APIs, advanced security enhancements, and performance optimization for large-scale deployments.