About 3 months after releasing this, Trek10 got spacklepunched by AWS when they launched when DynamoDB launched Auto Scaling natively. See here. This wasn't the first and won't be the last time, which is why we love working in such a fast-paced ecosystem.
The original post is below, but we highly recommend you ignore it and use native autoscaling!
Tue, 28 Mar 2017
We have several projects that use DynamoDB heavily. And, while DynamoDB works great, you can hit throttle limits if you are trying to tightly match your provisioned capacity to your actual demand. What about scaling up during the day and down during the night to meet your need? To do this, you would essentially need DynamoDB Auto Scaling.
We previously talked about how we do that at Trek10, but recently, David Wells from Serverless called us out during Serverless Office Hours and asked for it to be open sourced. I do want to point out that this implementation differs from our described methodology to using a simple polling Lambda function rather than CloudWatch alarms. This proved to be just a bit easier to manage configuration across many tables and indexes.
So David, this one is for you!
Repo: https://github.com/trek10inc/dynamodb-autoscaling Star