pragupta / wmlce-schematic

This repository holds a Terraform Schematic for WML-CE for use with the IBM cloud.

Deploy Watson Machine Learning Community Edition (WMLCE) software stack using Terraform on IBM Cloud VPC

Deploy a Linux Apache WMLCE stack on Ubuntu 18.04 ppc64le running in IBM Cloud VPC. These terraform scripts are also compatible with IBM Cloud Schematics.

Prerequisites

1. Terraform >= 0.11.14
2. IBM Cloud Terraform Provider >= 0.20.0
3. IBM Cloud API Key with authorization to provision in IBM Cloud VPC and IBM Cloud Block Storage for VPC

Overview

This deployment provisions:

1. a Linux Ubuntu ppc64le VM server on IBM Cloud VPC
2. Creates a new ssh key to login
3. Opens Ports 80, 443 in VPC Security Group port to access web page
4. Opens Ports 22 to access SSH console
5. Sets up NVIDIA driver and installs WMLCE packages

After install, you can ssh into the VM through the floating IP and private ssh key

To run the example, you will need to:

1. Clone this Git repository
2. Download and configure the IBM Cloud Terraform provider (minimally v0.20.0 or later)
3. Obtain your IBM Cloud API key (needed for step #4)
4. Update the variables.tf file to suit your needs

Provision Environment in IBM Cloud with terraform

Next, you can run the example by invoking...

The planning phase (validates the Terraform configuration)

terraform init
terraform plan

The apply phase (provisions the infrastructure)

terraform apply

The destroy phase (deletes the infrastructure)

terraform destroy

Future Improvements being worked on

• Not having to create a new Virtual Server Instance each time