We close Chapter 26 (and Part VI) with the most fundamental piece of any disaster recovery strategy: backups. As we saw in subchapter 26.2, even the most basic strategy (Backup & Restore) relies on having good backups. But doing backups well, for many different services, reliably and automatically, is not trivial. To solve this, AWS offers AWS Backup: a service that centralizes and automates backups for your entire infrastructure.

A typical company on AWS has data in many places: RDS databases (Chapter 8), server disks (EBS), files in EFS, data in DynamoDB... Each service has its own way of doing backups. Managing them separately is a mess and a risk:

The problems with doing it this way: it's easy to forget to configure backups for some resource (and discover it when it's too late), each service is managed differently, it's hard to verify that everything is being backed up, and there's no common policy. You need a single place to manage all backups.

AWS Backup is a service that centralizes and automates backups for many AWS services from one place. You define your backup rules once, and AWS Backup takes care of backing up all the indicated resources, automatically and consistently.

Analogy: AWS Backup is like hiring a professional backup service for your whole house, instead of each room taking care of itself. Instead of you remembering to back up your computer, your phone, your documents... a single service takes care of everything, according to a plan you define once, and guarantees that nothing is left without a backup. Centralized, automatic, and reliable.

You define backup plans that specify what to back up, how often, and how long to keep it, and they are applied automatically:

Notice how this connects with RPO (subchapter 26.1): the frequency of backups (how often) determines how much data you can lose. If you back up every hour, your RPO is one hour.

AWS Backup can apply a plan to all resources with a certain tag (remember tags, subchapter 6.x). So, any new resource you create with the "production" tag is automatically included in the backups, without you having to remember. This avoids the error of forgetting to back something up.

It automatically manages how long backups are kept and when old ones are deleted (to avoid infinite costs), and can even move old backups to cheaper storage (remember storage classes, subchapter 5.x).

From a single dashboard, you can see what is being backed up and what is not, making it easier to demonstrate (to auditors, for compliance) that your data is protected. This ties in with the compliance we saw in Chapter 23.

The great value of AWS Backup is the peace of mind of knowing that all your important data is backed up automatically, consistently, and verifiably, without depending on someone remembering to configure it resource by resource. Backups are your ultimate safety net: if everything else fails (accidental deletion, ransomware attack, catastrophic error), you can recover from the backups.

⚠️ A backup that hasn't been tested is not a backup. As important as making backups is verifying that they can be restored. Many companies discover their backups were corrupt or incomplete... just when they need them. Test your restores regularly.

Real-world example: a company suffers a ransomware attack that encrypts and damages data in several of its databases. Panic... until they remember they have AWS Backup with daily backups of all production resources (applied automatically by tag). They restore the databases to their state from the previous night. They lose, at most, a few hours of data (their RPO), but recover the business in hours instead of losing everything. The investment in centralized and automatic backups saved them. Without AWS Backup, maybe some critical database wouldn't have had a backup, with catastrophic consequences.

AWS Backup is the foundation on which disaster recovery strategies rely:

Without good backups, no recovery strategy works. AWS Backup ensures that this foundation is solid.

• Managing backups for each service separately is a mess and a risk (easy to forget something, inconsistent, hard to audit). You need to centralize.
• AWS Backup centralizes and automates backups for many services (RDS, EBS, DynamoDB, EFS...) from one place. Like a professional backup service for the whole house.
• Features: backup plans (what, how often, how long to keep), tag-based application (new resources are included automatically), lifecycle/retention management, and centralized auditing.
• The frequency of backups determines your RPO (subchapter 26.1): more frequent backups = less data lost.
• Provides the peace of mind of having all data automatically protected; backups are the ultimate safety net against deletions, attacks, or disasters.
• ⚠️ Test your restores: a backup that hasn't been tested is not reliable.

You have completed Chapter 26 and, with it, all of Part VI (Advanced AWS: security, observability, and costs)! You now master how to secure, observe, optimize, and protect your infrastructure. In Part VII we will take a leap into reference architectures and expert patterns, starting with the AWS Well-Architected Framework.