[SAA + SAP] 29. Disaster Recovery

Two Terms

• RPO: Recovery Point Object
• RTO: Recovery Time Object

Disaster Recovery Strategies

• Backup and Restore
• Pilot Light
• Warm Standby
• Hot Site / Multi Site Approach

• You can backup your on-premise data into S3 / Glacier, for example, once a week through Snowball, Storgage Gateway or internet
• EBS, RDS, Redshift can be also backup into S3 as Snapshot
• For recovery: 
  • Create EC2 from AMI
  • Get Snapshot of EBS from S3
  • Restore snapshot of RDS
• High RTO

• Keep most critical part of your application running, for example, database
• Do sync data replication to RDS on the cloud
• Not infrasctures are not running in normal mode (EC2, ELB)
• Recovery: 
  • Using Route 53, route to new EC2

• Setup a full copy of an application
• Just set min capacity for it
• Recovery:
  • Using Route 53 to point to standby ELB and modify its capacity

• Copy PROD application the same way as standby
• Fastest RTO
• Using Route 53 to switch DNS

• RDS replciation cross Region

Some applications might have an additional requirement to deploy their components using multiple regions; this can be a business or regulatory requirement. Any of the preceding scenarios in this whitepaper can be deployed using separate AWS regions. The advantagesfor both production and DR scenariosinclude the following: 

• You don’t need to negotiate contracts with another provider in another region 
• You can use the same underlying AWS technologies across regions 
• You can use the same tools or APIs