In this tutorial, we provide step-by-step instructions to go from loading a pre-trained Convolutional Neural Network model to creating a containerized web application that is hosted on Kubernetes cluster with GPUs on Azure Container Service (AKS). AKS makes it quick and easy to deploy and manage containerized applications without much expertise in managing Kubernetes environment. It eliminates complexity and operational overhead of maintaining the cluster by provisioning, upgrading, and scaling resources on demand, without taking the applications offline. AKS reduces the cost and complexity of using a Kubernetes cluster by managing the master nodes for which the user does no incur a cost. Azure Container Service has been available for a while and similar approach was provided in a previous tutorial to deploy a deep learning framework on Marathon cluster with CPUs .

The goal of Azure Container Services (ACS) is to provide a container hosting environment by using popular open-source tools and technologies. Like all software, deploying machine learning (ML) models can be tricky due to the plethora of libraries used and their dependencies. In this tutorial, we will demonstrate how to deploy a pre-trained deep learning model using ACS. ACS enables the user to configure, construct and manage a cluster of virtual machines preconfigured to run containerized applications. Once the cluster is setup, DC/OS is used for scheduling and orchestration.

Back in September, we announced Azure CLI 2.0 Preview. These commands provide a rich interface for a large array of use cases, from disk and extension management to container cluster creation. Today's announcement means that customers can now use these commands in production, with full support by Microsoft both through our Azure support channels or GitHub. We don't expect breaking changes for these commands in new releases of Azure CLI 2.0. This new version of Azure CLI should feel much more native to developers who are familiar with command line experiences in the bash enviornment for Linux and macOS with simple commands that have smart defaults for most common operations and that support tab completion and pipe-able outputs for interacting with other text-parsing tools like grep, cut, jq and the popular JMESpath query syntax .