YeeZTech/YeeZ-Privacy-Computing
FreeFidelius - YeeZ Privacy Computing 基于可信执行环境的熠智隐私计算中间件
FreeFree tier
About YeeZTech/YeeZ-Privacy-Computing
Fidelius (YeeZ Privacy Computing) is an open-source middleware that enables privacy-preserving data collaboration using Intel SGX trusted execution environments (TEE). It follows a 'data available but not visible' principle, ensuring original data never leaves the data provider's middleware instance. The system supports decentralized verification through blockchain, maintaining consistency of original data, controllability of computing logic, correctness of results, and privacy protection. Fidelius runs on Intel SGX hardware, with a debug mode available for environments without the required CPU.
Key Features
All-in-one privacy protection for data collaboration based on 'data available but not visible'
Ensures consistency of original data, controllability of computing logic, correctness of results, and privacy protection
Original data never leaves the data provider's middleware instance
Supports blockchain networks as trusted third-party for transmission and verification
Runs on Intel SGX trusted execution environments (TEE)
Debug mode available without SGX hardware by installing Intel SGX SDK
Open source with Apache 2.0 (library) and GPL 3.0 (binaries) licenses
Pros & Cons
Pros
- Open source with permissive licenses (Apache 2.0 for library)
- Hardware-based security via Intel SGX enclaves
- Data never leaves the provider, reducing leakage risk
- Blockchain integration adds verifiability and decentralized trust
- Community support and documentation available (Chinese and English)
Cons
- Requires Intel SGX hardware for full security guarantees (not available on all systems)
- Debug mode only works without SGX hardware, limiting production-level protection
- Complex setup involving BIOS, drivers, and Intel SGX SDK
- Limited to platforms supporting Intel SGX (no ARM or AMD equivalent supported)
Best For
Secure data collaboration between data providers and data users without exposing raw dataPrivacy-preserving joint analysis across organizations using sensitive datasetsVerifiable computation where results are guaranteed correct and private via blockchain attestationCompliance with data privacy regulations by ensuring data never leaves provider infrastructure
FAQ
What is Fidelius?
Fidelius is an open-source privacy computing middleware developed by YeeZTech that uses Intel SGX trusted execution environments to enable secure data collaboration while keeping raw data private.
What does 'data available but not visible' mean?
It means data providers make their data available for computation but the raw data is never exposed to the data user or any third party – only the computed results are shared.
Can Fidelius work without Intel SGX hardware?
Yes, a debug version can run after installing the Intel SGX SDK without requiring SGX-capable hardware, but this mode does not provide the same security guarantees.
How does blockchain integrate with Fidelius?
Fidelius can use a blockchain network as a trusted third party for secure transmission channels and verification of computation results, leveraging blockchain's decentralization and public verifiability.