⚡️ Serverless Plugin for DynamoDB Auto Scaling

With this plugin for serverless, you can enable DynamoDB Auto Scaling for tables and Global Secondary Indexes easily in your serverless.yml configuration file. The plugin supports multiple tables and indexes, as well as separate configuration for read and write capacities using Amazon's native DynamoDB Auto Scaling.

Usage

Add the NPM package to your project:

# Via yarn
$ yarn add sls-dynamodb-autoscaling

# Via npm
$ npm install sls-dynamodb-autoscaling

Add the plugin to your serverless.yml:

plugins:
  - sls-dynamodb-autoscaling

Configuration

Configure DynamoDB Auto Scaling in serverless.yml with references to your DynamoDB CloudFormation resources for the table property. The index configuration is optional to apply Auto Scaling Global Secondary Index.

custom:
  capacities:
    - table: CustomTable  # DynamoDB Resource
      index:              # List or single index name
        - custom-index-name
      read:
        minimum: 5        # Minimum read capacity
        maximum: 1000     # Maximum read capacity
        usage: 0.75       # Targeted usage percentage
      write:
        minimum: 40       # Minimum write capacity
        maximum: 200      # Maximum write capacity
        usage: 0.5        # Targeted usage percentage

That's it! With the next deployment, serverless will add a CloudFormation configuration to enable Auto Scaling for the DynamoDB resources CustomTable and its Global Secondary Index called custom-index-name.

You must provide at least a configuration for read or write to enable Auto Scaling!

Defaults

maximum: 200
minimum: 5
usage: 0.75

Index

If you only want to enable Auto Scaling for the index, use indexOnly: true to skip Auto Scaling for the general DynamoDB table.

API Throtteling

CloudWatch has very strict API rate limits! If you plan to configure Auto Scaling for multiple DynamoDB tables or Global Secondary Indexes, request an increase of the rate limits first! Otherwise, you might run into an error like this:

An error occurred while provisioning your stack: XYZ - Unable to create alarms for scaling policy XYZ due to reason: 
Rate exceeded (Service: AmazonCloudWatch; Status Code: 400; Error Code: Throttling; Request ID: XYZ).

Breaking Changes

There have been multiple breaking changes regarding CloudFormation resource names in the past. If you end up with an error, that your CloudFormation Stack cannot be updated, try to remove the custom > capacities configuration from your serverless.yml file and deploy the service without any Auto Scaling configuration. After that, just re-add your previous configuration and deploy your service again.

DynamoDB

The example serverless configuration above works fine for a DynamoDB table CloudFormation resource like this:

resources:
  Resources:
    CustomTable:
      Type: AWS::DynamoDB::Table
      Properties:
        TableName: custom-table
        AttributeDefinitions:
          - AttributeName: key
            AttributeType: S
        KeySchema:
          - AttributeName: key
            KeyType: HASH
        ProvisionedThroughput:
          ReadCapacityUnits: 5
          WriteCapacityUnits: 5
        GlobalSecondaryIndexes:
          - IndexName: custom-index-name
            KeySchema:
              - AttributeName: key
                KeyType: HASH
            Projection:
              ProjectionType: ALL
            ProvisionedThroughput:
              ReadCapacityUnits: 5
              WriteCapacityUnits: 5

License

Feel free to use the code, it's released using the MIT license.

Thanks

• TrentBartlem's Gist
• daniel's post in the AWS Developer Forums

Contribution

You are welcome to contribute to this project! 😘

To make sure you have a pleasant experience, please read the code of conduct. It outlines core values and beliefs and will make working together a happier experience.

Local Development

If you plan to change the TypeScript files and link the package to another project, use the link and build yarn commands:

$ > yarn link
$ > yarn build:watch # or "yarn build"

Use your local build of the package in another project:

$ > yarn link sls-dynamodb-autoscaling