Control your AWS cloud with a simple API that does all the undifferentiated heavy lifting for you.
Instead of having to learn the arcane details of each of the hundreds of AWS services, Control helps you think in terms of Apps, Services, Tasks and Resources, conceptually similar to Heroku or other PaaS offerings.
Run control setup to prepare your AWS account to host your apps & services. Internally, Control will configure a VPC, make private and public subnets in every availability zone, and set up the internet gateways, route tables and ACLs.
Run control create app to start a collection of Services and Tasks that make up a single App. Each App is a combination of a container and one or more starting commands — a codebase can have commands that start Sevices that work over HTTP, and Tasks that either run periodically via cron or 24/7. Internally, Control will set up an ECR repository, and configure your one or more of ECS, Fargate, App Runner, Cloud Map, HTTP Gateway, Lambda and ALB based on sensible defaults that you can override. All configurations are enabled for autoscaling, logging and metrics, with locked-down security groups.
Run control create database to set up a database of your choice on RDS, with sensible defaults for backups, IOPS and locked-down security groups.
Run control attach databaseID ––to appID to set up permissions for the app security group to reach the database security group, and add the DATABASE_URL environment variable to your application configuration.
Run control create topic to create a messaging topic, which gives you a fully integrated SNS + SQS combo that allows you to process any volume of messages easily. Internally, You can then do control attach topic --publisher publishingAppID or control attach topic --subscriber subscribingAppID to add the SNS topic URL or SQS queue URL to the application, with the correct IAM permissions set to send or receive & delete messages.
control works with great with AWS CloudShell, helping you control your apps and services from anywhere without having to manage credentials. It also runs fine on your local machine, using the AWS credentials configured according to the default conventions.
All control commands support JSON output for easy scripting.