Automating Cloud Deployments

Ned Bellavance described cloud deployment automation using infrastructure-as-code tools. His presentation included:

• Infrastructure-as-code concepts and tools;
• Terraform basics;
• Using Terraform with Azure;
• Terraform and server configuration management;
• Automating Terraform

Most traditional state management and tools and automation frameworks can be used to automate cloud deployments.

This section briefly describes the architecture of three most common automation tools (Chef, Puppet, and Ansible), and introduces the concepts of version control and reviews.

Git is the source-control tool of choice not only for most open source project, but also for extremely large teams like Facebook or Microsoft Windows development team. It seems a bit convoluted when you start, but you’ll quickly discover its benefits in this Getting started with Git tutorial by Scott Lowe.

If you’re familiar with another version control tool, or want to learn more about Git, I’d highly recommend reading the Pro Git book. I would also suggest you invest more time in understanding Git concepts than trying to remember a few quick recipes and move on.

You could use PowerShell instead of Linux CLI to automate Azure deployments.

If you want to go down that path, and are not familiar with PowerShell, you might find this introductory presentation useful.

Ansible includes numerous AWS, Azure and GCP management modules, and if you plan to use it in your other automation projects you might as well try to use it to automate your cloud deployments (although you might find other tools like Terraform easier to use).

Access content

In this presentation David Barroso focused on two important automation aspects:

• How to manage and consume secrets safely in our automation platform; from traditional methods like encrypted files to modern systems like Hashicorp Vault
• How to leverage modern platforms to dynamically provision certificates and SSH keys so we can keep communications with our devices secure.

AWS API has numerous idiosyncrasies: it's a strict Create/Read/Update/Delete API, you need multiple API calls to create an object, objects can be referenced by IDs, but not by names...

This section explains the peculiarities of AWS API, and describes how to use bash scripts, Ansible playbooks, CloudFormation templates, and Terraform configuration files to create, update, or delete anything from a single AWS object to a full-blown application stack.

It's a pleasure to work with Azure API: objects have names, you can refer to them by names, and the API uses create-or-update approach, making automation scripts or playbooks extremely easy to develop or understand.

This section explains the principles of Azure API, and describes how to use bash scripts, Ansible playbooks, Terraform configuration files, and Resource Manager Tempaltes to create, update, or delete networking- and compute infrastructure.

In this assignment you'll select a public cloud provider you'd like to work with, create your public cloud account, select an IaC tool of your choice, and create some simple public cloud infrastructure with that tool.

This section contains links to external blog posts, podcasts... that you might find useful after mastering the basics of public cloud automation.