The seven secrets to high cloud availability
As more companies embrace the mixed approach of hybrid IT – retaining some IT infrastructure internally and moving some to the cloud – it is becoming increasingly important for companies to ensure they have high-availability cloud infrastructure. There are several ways to achieve this.
“When systems are migrated to cloud environments, they often need to be moved in stages while remaining aligned with internal infrastructure. At the same time, traditional systems are increasingly being augmented by cloud-based applications. This is where hybrid IT becomes necessary,” says CenturyLink regional director Stuart Mills.
“Application reliability and availability can become much more complex to ensure in a hybrid IT environment. By modelling the resultant hybrid system using techniques that are commonly used in distributed computing, overall availability can be improved for assured uptime and service levels.”
There are seven secrets to ensuring the high availability of hybrid applications:
1. Use a managed DNS service
A highly-reliable managed Domain Name System (DNS) service will enable dynamic traffic re-routing when application components become unavailable for any reason.
2. Assess failure risks
For a hybrid IT deployment scenario, applications should be designed with failure scenarios in place. A best practice is to evaluate the reliability of networks, machines, and software for each location involved.
3. Calculate availability
Hybrid applications that run on multiple platforms are inherently distributed, and the availability of each individual platform can be used to calculate the end-to-end application availability.
4. Balance loads
So-called load balancers can be used to route data traffic between multiple application server and web server instances. These redundant web and application servers balance data loads and provide both additional capacity and failover.
5. Use horizontal auto-scaling
Horizontal auto-scaling involves adding or removing virtual servers. With horizontal auto-scaling, applications can respond to changes in demand or load within minutes.
6. Use vertical auto-scaling
Vertical auto-scaling can add power to existing servers instead of adding additional servers, as is the case with horizontal auto-scaling. The choice between horizontal and vertical auto-scaling will depend on the specific use case.
7. Manage storage for high availability
Storage is another major factor that affects high availability in a hybrid environment. Best practice is to make sure that cloud applications have storage that’s explicitly architected for fault-tolerance.