AWS services

AWS S3
- daily, encrypted backups of MongoDB replica set
- raw device upload data
- logs of all access to AWS resources via CloudTrail, CloudWatch, Config
- Note: S3 is not region-specific. Even if US-West 2 goes down entirely (a catastrophic event that would cause major Internet service disruption), S3 is still available. This would allow Tidepool to deploy a new production environment in a new region in a matter of hours.
AWS Simple Email Service (SES)
- Used to send all Tidepool service emails (e.g., new account verification, forgot password)
- Two dedicated SES instances assist in security/logging of email/email monitoring
AWS DynamoDB
- Used for saving application metrics and Analytics development
AWS Route53: provides Tidepool DNS
AWS CloudTrail: records AWS API calls
AWS CloudWatch: provides monitoring for AWS cloud resources and applications
AWS CloudFormation: management and provisioning of AWS resources
AWS Config: resource inventory, configuration history, and configuration change notifications for security and governance
AWS EKS: managed Kubernetes (k8s) services
AWS Secrets Manager: handling/storing/providing sensitive configuration data
AWS GuardDuty: network security monitoring
AWS ELB: Amazon Elastic Load Balancing
AWS SNS: Amazon Simple Notification Service - triggered relay of logs and notifications
AWS ASG: Auto Scaling Groups assist in maintaining and dynamically adjusting desired number of systems/resources
AWS VPC: Virtual Private Cloud logically separates and defines network and compute environment resources

All environments are configured identically to PRD, with these differences:

Separate networking environment, database and compute
Reduced resource demands (smaller environment)

All of the instances in each environment are self-contained in AWS Virtual Private Cloud (VPC) to allow network isolation so a change in Development has no way to affect anything in Production.

There are also a variety of support infrastructure components for each environment (load balancers, auto-scaling, public/private networks, monitoring and logging, metrics, etc.). These resources are accessed by the environments as common services in a separate cluster.