← AWS SAA

Note for Disaster Recovery

awscloud

RPO and RTO

RPO stands for Recovery Point Objective.
RTO stands for Recovery Time Objective.
RPO is the backup point before the disaster.
RTO is the recovery point after the disaster.

Disaster Recovery Strategies

Backup and Restore(High RPO and RTO)

Backup and store on somewhere.
Restore back after the disaster.

Pilot Light

A small version of app running.
When the disaster occurred the DNS resolver re-route to the small version of app when the main one is down.
Faster than Backup and Restore as the critical systems are already up.

Warm Standby

Full System up and running but with minimum resources.
Upon disaster, that system can be increased load to run as the production system

Hot Site/Multi Site Approach

Very Low RTO (expensive)
Same production level system is running on both AWS and On-premise
When disaster occurred, the DNS resolver will route only to the active one.
This is called Active - Active

Elastic Disaster Recovery

Used to recover virtual, on premise and other cloud servers to AWS

Database Migration Service

Used to migrate databases to AWS.
Source database remains available during migration.
Support
Homogeneous: eg. Oracle to Oracle
Heterogeneous: eg. MSQL server to Aurora

Schema Conversion Tool

If the source and target don't use the same engine, We have to use SCT.

RDS to Aurora migration

From AWS MySQL RDS to Aurora

From External MYSQL to Aurora

AWS Backup

Use as a central service to automate backup AWS services.
Support cross-region, cross-account backups.
usage
create backup plan -> assign to AWS services

Backup Vault lock

Additional layer to defense backup
WORM(Write Once Read Many) state for all backup stores in AWS backup vault

Application Migration Services

Discovery Service

Used to plan the migration by gather information about the on premise data center.

Migration Service(MGN)

Lift and shift.
Convert physical, virtual and cloud-based server to run natively on the AWS.

Transferring Large Amount of Data to AWS

Sorted in duration that need to be finished

For on-going replication/transfers: Site to Site VPN or DX with DMS or DataSync

VMWare Cloud on AWS

VMware Cloud on AWS lets you run VMware workloads on AWS infrastructure without re-architecting them. It's a managed service, so AWS handles the underlying hardware and vSphere/vSAN/NSX stack. You can migrate existing on-prem VMs to the cloud using VMware HCX with minimal downtime. It also works well for disaster recovery, spinning up a DR site on AWS instead of maintaining a second data center. Overall it bridges your existing VMware environment with native AWS services.