You seem to have skipped mentioning all the things that are clearly superior in GCP vs AWS
Instance names are hostnames, boot up a machine named "salt" and it will show up as "salt.c.project-name.internal" as well as "salt" on all your machines making several things VERY easy compared to AWS. This internal DNS is very easy to expose over VPN etc. as well. Fully automated.
Startup scripts are project-wide. How is this not a thing in AWS in 2017?
You use YOUR SSH keys to connect to instances, and they can be set up project-wide if you so wish. You don't use some random generated keys that you then have to figure out how to store securely and distribute in addition to your already existing perfectly good SSH keys.
Since hostnames are correctly set up based on the instance name from the start it's really easy to do instance matching in various tools, such as in Salt Stack, without a need for some random AWS integrations. Just match api-* for all api servers, etc.
Instance groups and templates make setting up an easy to scale (including automatically) group of machines that will distribute instances across availability zones easily.
The load balancer is Google Load Balancer, not some random instance that will die when you get hit with a traffic spike. You turn on the LB in GCP and it will instantly handle a million requests per second.
Overall the whole interface and every bit about their system has been built to be easier and more powerful to use than AWS
There is no wondering about provisioned IOPS and all that nonsense. Size = performance in GCP. 500GB disk is 2x as fast as 250GB disk.
Long story short, having used GCP for the past 3-4 years or so having to do a few things with AWS recently was a huge blast from the past and it was hard to believe how bad things can still be in AWS.
GCP is easier to use, more powerful, less prone to human error, cheaper, and generally just superior in every single way.
Other than the SSH keys, I haven't touched any of this, so thanks for chiming in. :)
I'll probably be using some of that in future experiments!
We’re a place where coders share, stay up-to-date and grow their careers.
We strive for transparency and don't collect excess data.