Scan AWS resources for idle periods

AWS Compute Optimizer helps us to prune our unused infrastructure by identifying idle resources - what about when resources are necessary but are idle periodically.

That's why we built upscan - a simple CLI tool to identify consistently idle periods for EC2/ECS/RDS/NAT Gateways and recommend an on/off schedule.

Why

Driven by our teams desire to be cost-efficient with our non-prod environments, we needed a way to confidently identify our usage patterns for AWS infrastructure.

As a result we managed to reduce our non-prod costs by 70%.

What does it look like

How

A simple homebrew install:

brew install uptime-scheduler/tap/upscan 

Log in to AWS via the CLI and configuring your session (upscan never sees your credentials).

Then run something like:

upscan --region us-east-1

If terminal outputs aren't for you or you want to share results with your team or management you can generate HTML reports with:

upscan --region us-east-1 --output html --save report.html

What do you think?

We would love to hear any and all feedback. Was this useful? What features are missing? Let us know.

Stay safe out there with AWS cost creep/explosion!