SGX-PySpark: Secure Distributed Data Analytics presented at The Web Conference 2019

Researcher Do Le Quoc, a postdoctoral fellow in the Systems Engineering Group of TU Dresden, showcased his work on the LEGaTO project at The Web Conference 2019 held in San Francisco, USA on May 13-17. He presented a live demo at the Pacific Concourse area of the event at 2:00 pm on May 16. He also exhibited a poster and answered questions from interested participants.

In his work, “SGX-PySpark: Secure Distributed Data Analytics”, Do addressed how public cloud users can protect sensitive data while still preserving the same utility of data analytics.  In his live demo, he showed how to perform data analysis in an untrusted cloud in a secure manner using PySpark and Intel SGX with the help of the SCONE platform. In this approach, both input data and data analytics codes are encrypted before uploading to the cloud infrastructure. Thereafter, in the cloud, they are securely decrypted and executed inside Intel SGX enclaves. The proposed framework can be used to protect sensitive data such as healthcare records, banking information, as well as other personal information.