This is the first of many posts in a series I am embarking chronicling my work on using Google Cloud Platform (GCP) for creating Big Data applications ? Why Google Cloud and not AWS you may ask ? Well, I already use AWS extensively at work, so at home I’m deciding to do something different so I can broaden my expertise. On a secondary note, it seems as if GCP may be a cheaper alternative than AWS for compute intensive workloads – and that matters when you pay for compute resources out of pocket.
So I already had a GCP account I had created a few years back. My first task was to figure out how to create VM instances on GC via the command line API gcloud and then via Ansible.
Via Command Line API
I followed the instructions for setting up the gcloud client on my Ubuntu laptop.
I was subsequently able to create an instance with the following command:
I was prompted to select a region and I chose us-east-1
gcloud compute instances create test-instance --image-family ubuntu-1710 --image-project ubuntu-os-cloud
My goal is fully automating the provisioning of resources in GCP, so the next step for me would be to figure out how to provision a VM instance using cloud automation software such as Puppet, Ansible.
Ansible is what I am most familiar with from work so Ansible it is.
The simplest and quickest way to get started is by reading the example in the Ansible Google Cloud Platform Guide.
The video is instructive, and subsequently I cloned the repo and attempted to follow the instructions to create my instance. I was able to create instances via the following :
and subsequently terminate them as follows: