In a two-zone cloud DR configuration, Zone “A” and Zone “B”, each zone replicates the other such that if a zone experiences a downtime as a result of disaster, then the other becomes active.
Each zone has its own internal System Architecture (SA) similar to the other.
The zone “A” has an Active Load Balancer and zone “B” has a Passive Load Balancer as shown in the figure below.
The components of the cloud disaster recovery architecture are discussed as follows.
Active Load Balancer
The primary function of a load balancer is to ensure fair distribution of the system’s work load. The load balancer is also responsible for citizen request identification and evaluation to include the type and volume or requests and confirms if the available resources can serve these requests or not.
If the available resources are not adequate to manage the citizen’s request, then it signals the monitoring unit to provision additional resources to the system architecture.
This reflects the scalability features of cloud computing. Active Load Balancer also synchronizes with SA in zone “B”.
Passive Load Balancer
This is the same as Active Load Balancer except that is it in the passive mode. It becomes active within 5 to 10 minutes when zone “A” experiences a downtime due to a disaster.
This unit manages the activities and states (good/compromised) of both zone “A” and zone “B” to improve efficiency and reliability of service.
It gets resource constraints’ feedback from the load balancer and verifies the request. If additional resource(s) are truly required, it responds with the appropriate resource back to the load balancer to execute accordingly.
It is responsible for the switch in the active role between zone “A” and zone “B”. depending on their internal states. If zone “A” experiences a disaster or found to have been compromised, then it is taken down and zone “B” is made active and vice versa.