Architecting the Agile Enterprise with the Agile Cloud Manager
Part 14 of 16: Fail-Over and Fail-Back Between Any Cloud, On-Prem, or Edge
The two diagrams in this section illustrate how designing each of the requirements for failover and failback can become a very simple process using the Agile Cloud Manager.
Figure 14 illustrates some of the aspects of designing for multiple clouds, on-prem, and edges using the Agile Cloud Manager.
A budget for multi-cloud can be created from savings that can be accrued from the efficiencies that result from the reduced engineering and operations costs that can result from using the Agile Cloud Manager.
Everything above the system template level can be virtually identical between cloud providers due to the Agile Cloud Manager’s Command Line Interface CLI’s ability to abstract away the inner details of systems.
System templates for the same system in different clouds will be almost identical because the Agile Cloud Manager’s Domain Specific Language DSL enables you to model the same system the same way for every cloud, with the same service types, the same service instances, and the same foundation.
The only thing that might be different between system templates for the same system in different clouds might be the variables needing to be mapped differently within each object to fit the underlying building blocks.
Basic building blocks for different clouds can be analogous to each other if you are careful to use identical operating systems and compatible network designs.
Figure 15 illustrates how simple pipeline design can be for Failover and Failback using the Agile Cloud Manager.
Failover and Failback between cloud, on-prem, and edge providers is a lot like doing Blue/Green deployments in the sense that your automation creates a complete cloned copy of your Production environment and then redirects traffic to the newly cloned Production environment. The differences are that:
- The new clone of Production is in a different cloud, different on-prem, or different edge.
- The application versions in the new Production are the same before and after a failover or a failback.
Data replication between the cloned appliances becomes a much more isolated and reasonably scoped engineering task.
Next: Proceed to Part 15 of 16
back to Site Home
back to Architecture section Home
Back to Part 13 of 16