Backing Up What You Run in AWS. The Cloud Is Not a Backup
Amazon looks after the infrastructure. Protecting your data from deletion, mistakes and attacks is still your job.
AWS describes a shared responsibility model. Amazon is responsible for the security and resilience of the underlying cloud, such as its buildings, hardware and networks. You are responsible for what you put in it, including your data, your configuration and who can access it. If an employee deletes a database, or an attacker gets hold of an administrator login, AWS does not automatically put things back.
AWS provides tools to help. AWS Backup is a managed service that lets you set backup plans for supported services such as EC2 volumes, databases and file systems from one place, with schedules and retention rules. Snapshots of EBS volumes and databases can be taken and kept for a defined time. S3 offers versioning, which keeps earlier copies of objects, and replication features to copy data to another location.
Where your backups live matters as much as having them. Copying backups to a separate AWS account, and where appropriate another region, protects you if the main account is compromised. AWS Backup also offers a locking feature, called Vault Lock, designed to stop backups being altered or deleted before their retention period ends. That is a strong defence against ransomware and against mistakes.
Do not skip testing. A backup you have never restored is an assumption. Pick a system, restore it into a separate environment, time how long it takes and note what was missing. This gives you a realistic idea of recovery time, which is what matters when the business is waiting.
Finally, remember that backup costs money to store, so set retention to match what you actually need and check the current AWS pricing. If you run workloads in AWS and want an honest review of how well they are protected, Dragonfly IT would be glad to help.
