Teaclave: a secure and easy-to-use privacy computing platform

Mingshen Sun

Chinese Session 2021-08-07 13:30 GMT+8  #incubator

In the era of big data, open data sharing, data value, data resource integration and security protection have become an important topic. Traditional data processing methods can no longer meet the current needs of data circulation, computing as well as security. Privacy protection technology has emerged to solve this problem, break down data barriers and solve the data silo dilemma.

Apache Teaclave (incubating) is a privacy-secure computing platform that empowers private data computing. Based on hardware security capabilities, Teaclave ensures the secure and controlled flow and processing of sensitive data in trusted out-of-domain and off-shore scenarios, without fear of privacy data leakage and misuse. Teaclave also supports multi-party federated computing, breaking down data silos in enterprises and organisations. teaclave was donated by Baidu to the Apache Foundation Incubator in 2019. The first open source community release is due in October 2020.

Teaclave uses Intel SGX and ARM TrustZone technologies to provide security techniques based on hardware isolation, memory encryption, remote authentication and other data privacy protection computing tasks. The Teaclave platform provides a function-as-a-service interface, lowering the barrier to use. The platform also implements many built-in functions, such as machine learning algorithms, multi-party federated modelling, encryption and decryption calculations, among other common functions. What’s more, developers can freely write functions in Python to manipulate private data and execute them in the platform. Finally, to avoid the security risks associated with memory security vulnerabilities, Teaclave is also written in the memory-safe programming language Rust.

In this talk, I will present the background of privacy-safe computing, the Teaclave system architecture and design philosophy, security enhancements, and finally open source community building.


Mingshen Sun: Mingshen Sun works at Baidu and is a member of Apache Teaclave (incubating) PPMC (Podling Project Management Committee). He leads, maintains and actively contributes to several open source projects including Teaclave FaaS Platform, SGX/TrustZone SDK, MesaPy, etc. Please visit his homepage (https://mssun.me) for more information.