Implementing data backups and replications

Task Statement 1.3: Determine appropriate data security controls.

📘AWS Certified Solutions Architect – (SAA-C03)


1. Introduction

Data is one of the most important assets in any IT system. If data is lost, corrupted, deleted, or encrypted by ransomware, the business can stop functioning.

For the SAA-C03 exam, you must understand:

  • What backups are
  • What replication is
  • The difference between backup and replication
  • Which AWS services provide backup and replication
  • How to design solutions for high availability and disaster recovery
  • Cost, performance, and security considerations

This topic is heavily tested in scenario-based questions.


2. What Is a Data Backup?

A backup is a copy of data stored separately so it can be restored later if the original data is lost or damaged.

Backup protects against:

  • Accidental deletion
  • Ransomware
  • Data corruption
  • Application bugs
  • Infrastructure failure
  • Insider threats

Backups are usually:

  • Stored in a different location
  • Stored in another Availability Zone (AZ)
  • Stored in another AWS Region
  • Kept for a defined retention period

3. What Is Replication?

Replication means copying data automatically from one system to another system, usually in near real time.

Replication is mainly used for:

  • High availability
  • Disaster recovery
  • Low latency access
  • Business continuity

4. Backup vs Replication (Important for Exam)

FeatureBackupReplication
PurposeRestore lost dataKeep systems running
SpeedScheduledNear real-time
Protects from deletion?YesUsually No
Protects from ransomware?Yes (if isolated)No (damage replicates)
Used for DR?YesYes
Storage costLowerHigher

Very Important Exam Point:

If data is deleted accidentally:

  • Replication will copy the deletion.
  • Backup will allow restoration.

5. AWS Services for Backups

You must know the major AWS services used for backups.


5.1 Amazon S3 Backup Features

S3 Versioning

  • Keeps multiple versions of an object.
  • Protects against accidental deletion or overwrite.
  • Must be enabled manually.

S3 Cross-Region Replication (CRR)

  • Automatically replicates objects to another Region.
  • Used for disaster recovery.

S3 Lifecycle Policies

  • Move data to cheaper storage classes:
    • S3 Standard
    • S3 Standard-IA
    • S3 Glacier
    • S3 Glacier Deep Archive

S3 Object Lock

  • Prevents deletion for a defined period.
  • Used for compliance (WORM – Write Once Read Many).

Exam Tip:
If question mentions protection against accidental deletion, answer is often S3 Versioning.


5.2 Amazon EBS Snapshots

  • Snapshots are incremental backups.
  • Stored in S3 (managed by AWS).
  • Can restore entire volumes.
  • Can copy snapshots to another Region.

Key Concepts:

  • First snapshot = full backup
  • Later snapshots = incremental
  • Can create AMIs from snapshots

Exam Tip:
If EC2 data must be backed up → Use EBS snapshots.


5.3 Amazon RDS Backups

RDS provides:

Automated Backups

  • Enabled by default
  • Point-in-time recovery (PITR)
  • Retention: 0–35 days

Manual Snapshots

  • Must be deleted manually
  • Used for long-term retention

Cross-Region Snapshot Copy

  • Used for disaster recovery

Important:
Multi-AZ is replication, not backup.


5.4 Amazon DynamoDB Backups

Two types:

On-Demand Backups

  • Full backup
  • Long-term storage

Point-in-Time Recovery (PITR)

  • Continuous backup
  • Restore to any second in last 35 days

5.5 AWS Backup

Centralized backup management for:

  • EBS
  • RDS
  • DynamoDB
  • EFS
  • FSx
  • EC2
  • Storage Gateway

Features:

  • Backup policies
  • Cross-account backups
  • Cross-Region backups
  • Lifecycle management
  • Encryption support

Exam Tip:
If question asks for centralized backup across multiple services → Answer is AWS Backup.


6. AWS Services for Replication


6.1 RDS Multi-AZ (High Availability)

  • Synchronous replication
  • Standby in another AZ
  • Automatic failover
  • Used for high availability, NOT read scaling

If question says:

  • “Automatic failover”
  • “High availability”
    Answer is Multi-AZ.

6.2 RDS Read Replicas

  • Asynchronous replication
  • Used for read scaling
  • Can be cross-Region
  • Can promote to standalone DB

6.3 S3 Replication

  • Same-Region Replication (SRR)
  • Cross-Region Replication (CRR)
  • Asynchronous

6.4 Amazon EFS Replication

  • Automatic cross-Region replication
  • Used for disaster recovery

6.5 AWS Elastic Disaster Recovery

  • Continuous replication of servers
  • Minimal downtime
  • Supports on-premises to AWS
  • Supports cross-Region recovery

Exam Scenario:
If company wants low RPO and low RTO for entire application → Use Elastic Disaster Recovery.


7. Disaster Recovery Concepts (Very Important)

You must understand:

RPO (Recovery Point Objective)

How much data loss is acceptable?

Example:
RPO = 5 minutes
Maximum 5 minutes data loss allowed.

RTO (Recovery Time Objective)

How fast system must recover?

Example:
RTO = 10 minutes
System must be restored within 10 minutes.


8. DR Strategies in AWS (Exam Favorite)

1. Backup and Restore

  • Lowest cost
  • Highest RTO
  • Suitable for non-critical workloads

2. Pilot Light

  • Core services running
  • Scale up during disaster

3. Warm Standby

  • Small but fully working environment
  • Faster recovery

4. Multi-Site (Active-Active)

  • Fully running in multiple Regions
  • Lowest RTO
  • Highest cost

Exam Tip:
If question says:

  • “Lowest cost” → Backup & Restore
  • “Minimal downtime” → Multi-site

9. Cross-Region vs Cross-AZ

Multi-AZ

  • Protects from AZ failure
  • Same Region

Cross-Region

  • Protects from Region failure
  • Used for disaster recovery

Important:
Compliance rules may require cross-Region backups.


10. Security Best Practices for Backups

For the exam, remember:

1. Encrypt Backups

Use KMS encryption.

2. Enable Versioning

Protects from deletion.

3. Use MFA Delete (S3)

Adds extra protection.

4. Cross-Account Backups

Protect against account compromise.

5. Restrict Access with IAM

Only authorized users can delete backups.

6. Test Restores

Backup without testing restore = risky design.


11. Common Exam Trap Scenarios

  1. “Database must be highly available”
    → Multi-AZ
  2. “Database must scale read traffic”
    → Read Replica
  3. “Data must be protected against accidental deletion”
    → Backup or Versioning
  4. “Centralized backup across multiple services”
    → AWS Backup
  5. “Lowest cost DR”
    → Backup and Restore
  6. “Minimal data loss”
    → Synchronous replication

12. Cost Considerations

Backups cost:

  • Storage
  • Data transfer (cross-Region)
  • Snapshot storage

Replication costs:

  • Additional infrastructure
  • Data transfer
  • Compute resources

Exam Tip:
Always choose most cost-effective solution that meets requirements.


13. Final Exam Summary – What You Must Remember

You must clearly understand:

  • Difference between backup and replication
  • RPO and RTO
  • Multi-AZ vs Read Replica
  • Snapshot vs Automated Backup
  • Versioning vs Replication
  • Cross-Region vs Cross-AZ
  • AWS Backup capabilities
  • Disaster recovery strategies
  • Encryption of backups

Final Advice for the Exam

When reading a scenario:

  1. Identify if it is about:
    • High availability
    • Disaster recovery
    • Data protection
    • Compliance
    • Cost optimization
  2. Check:
    • Required RPO
    • Required RTO
    • Budget constraints
  3. Then select:
    • Backup
    • Replication
    • Multi-AZ
    • Cross-Region
    • Or combination
Buy Me a Coffee