High Availability and Disaster Recovery
The architecture of high availability and disaster recovery (HADR) is as follows:
When an entire data center goes down due to natural disaster, equipment failure, or cyber attack, a business has to recover lost data from where the data is backed up. The disaster recovery relies upon the replication of the backed up data in a safe network or a cloud location that is not affected by the disaster.
Disaster recovery architecture can be setup using Cold standby mode or Warm standby mode.
The two most important parameters for a disaster recovery plan are:
- Recovery Point Objective (RPO). Describes the age of files that must be recovered from backup for a business operation to resume after a disaster. It also specifies how often you should back up data. For example, if your RPO value is 15 minutes, then the data before 15 minutes of a disaster must be restored for operations to resume.
- Recovery Time Objective (RTO). Describes the duration and service level within which you must restore the most critical IT services after a disaster. For example, if your RTO value is 60 minutes, the data in the required systems must be restored within 60 minutes of a disaster event.
You can have an effective disaster recovery management in place by configuring a reliable repository and by taking data backup at regular intervals.
- Failover. The failover operation is the process of switching data from a primary data center to a backup facility.
- Failback. The failback operation is the process of returning data from the backup facility to the primary data center.