v1.0.0

Intel® Cloud Optimization Module for Kubernetes* v1.0.0

The Intel® Cloud Optimization Module for Kubernetes* can be used to build and deploy AI applications on Microsoft Azure with Intel accelerations for machine learning workloads. Demonstrated in this module are Intel optimizations for training and inference of an XGBoost classification model for loan default risk prediction.

The architecture uses Docker for application containerization and stores the image in an Azure Container Registry (ACR). The application is then deployed on a cluster managed by Azure Kubernetes Service (AKS). Our clusters run on confidential computing virtual machines leveraging Intel Software Guard Extensions (SGX). We use a mounted Azure File Share for persistent data and model storage. The client interacts with our infrastructure through the Azure Load Balancer, which gets provisioned by our Kubernetes service.

Features

• Azure Kubernetes cluster leveraging Intel SGX confidential computing VM nodes
• Data processing pipeline
• Loan default risk prediction model with Intel optimizations for XGBoost and Daal4py
• Inference endpoint for real-time model predictions