Protecting privacy and confidentiality is critical for large-scale data analysis and machine learning. BigDL Privacy Preserving Machine Learning (PPML) provides a trusted cluster environment for secure big data and AI applications, even on an untrusted cloud environment.

By combining Intel® Software Guard Extensions with several security technologies (such as attestation, key management service, private set intersection, federated learning, and homomorphic encryption), BigDL PPML ensures end-to-end security that is enabled for entire distributed workflows (such as Apache Spark*, Apache Flink*, XGBoost, TensorFlow*, and PyTorch*) on Kubernetes* in a cloud native envrionment.