Abstract:

With the rise of data-driven material design driven by artificial intelligence and materials science, material science data has become a focal point of production factors, national strategic resources, and international competition. However, as material data sharing increases, data security issues become increasingly important. Issues such as data leakage, misuse, and tampering threaten the competitiveness of enterprises. We first review mainstream data security protection technologies, including access control and encryption technologies, which constitute the traditional data security protection model, ensuring security during data transmission and storage. Next, the development of blockchain technology is introduced. Blockchain technology can achieve confidentiality, integrity, and availability during data transmission and storage, but these mechanisms still cannot address privacy issues during data usage, nor can they protect the confidentiality and integrity of data during usage. Then, the advantages of confidential computing technology are analyzed. By executing calculations in a hardware-based trusted execution environment, confidential computing technology minimizes the trusted computing base, providing comprehensive data protection and adhering to the concept of "data usability without visibility" to protect data during usage, thereby constructing end-to-end lifecycle data security. Finally, we combine the advantages of blockchain and confidential computing technology to propose a trustworthy infrastructure solution for material data based on blockchain and confidential computing, to achieve security throughout the data lifecycle and provide strong support for the secure application of material data.

Key words: confidential calculations, blockchain, material data, data security, data sharing