Enhanced secure storage and data privacy management system for big data based on multilayer model.

Area of Science:

• Computer Science
• Data Security
• Cloud Computing

Background:

• Managing and securing large-scale sensitive data, particularly personnel records, is a significant challenge in cloud environments.
• The increasing complexity and scale of big data systems necessitate advanced security solutions.

Purpose of the Study:

• To propose a novel Multi-Layer Secure Cloud Storage Model (MLSCSM) for large-scale personnel data in cloud environments.
• To integrate cryptographic and statistical methods for privacy-preserving and secure data storage.

Main Methods:

• The MLSCSM combines ChaCha20 encryption, Dual Stage Data Partitioning (DSDP), k-anonymization, SHA-512 hashing, and Cauchy matrix-based dispersion.
• Data blocks are securely encoded, masked, and distributed across multiple cloud platforms based on various factors.
• The model incorporates audit logs, load balancing, and real-time resource evaluation.

Main Results:

• The proposed model achieved a 250 ms encoding time (block size 75), 23% CPU usage for 256 MB data, and low latency of 14 ms.
• Demonstrated a high throughput of up to 139 ms, outperforming baseline models like RDFA, SDPMC, and P&XE.
• Validation using the MIMIC-III dataset on a Hadoop cluster confirmed the system's effectiveness.

Conclusions:

• The MLSCSM offers superior security, efficiency, and scalability for cloud-based big data storage applications.
• The integration of cryptographic and statistical techniques provides a robust solution for privacy-preserving data management.
• The model is optimized for distributed Cloud Computing Environments (CCE).