Toward secure and efficient data analytics in serverless computing

Abstract

In this paper, we propose a secure and efficient data analysis system in serverless computing using Intel SGX. We utilize Intel SGX to preserve the confidentiality and integrity of user data in a cloud-untrusted environment. Although various studies have been proposed using Intel SGX, there is a lack of research on detailed key exchange protocols using remote attestation, and there is also a lack of research on the execution of encrypted code by real users. In this paper, we propose a key exchange protocol based on Intel EPID remote attestation and a dynamic loading method utilizing SGXv2. In addition, we propose a data encryption format for efficient shuffling process. By building an environment using containers similar to the actual serverless computing environment, we show the possibility to actual serverless computing. Finally, we discuss the limitations of this research and suggest future research directions.