Disaster recovery using the cloud

Did you like our article?

Public cloud services may be used to strengthen the Disaster Recovery (DR) strategy of a company’s critical services in the event of a catastrophe.

The Disaster Recovery Plan (DRP) is a procedure to protect the company’s mission-critical business processes in the event of an unrecoverable disaster in the primary datacenter.

One of the elements to consider in a Disaster Recovery Plan is to prepare one or more datacenters to launch critical services in the event of the primary datacenter’s total unavailability, especially when recovery is not possible for a certain period of time, in order to avoid disturbing daily operations.

A well-designed DR strategy must allow the company to operate will few or no drawbacks in a disaster scenario.  One of the latest trends when defining DR solutions is the incorporation of public cloud services to host the contingency datacenter(s).

As a premise, one of the main advantages of having infrastructure in public clouds are the cost and management flexibility these services offer, as well as the easiness to build high-availability architectures, and in the case of Disaster Recovery this is not an exception.

Another important benefit of having a Disaster Recovery solution in a public cloud is protecting the company’s processes on the assumption a disaster could render all the existing datacenters in the country unavailable. In this scenario, the company could continue to operate its business processes in the cloud.
In addition, a DR on public clouds allows for cost flexibility versus recovery time (Recovery Time Objective, RTO), and versus how many minutes of data can be lost (Recovery Point Objective, RPO).
A common example is to have the cloud DR datacenter identical in size to that in production, allowing low recovery times when facing a disaster, and aiming always towards having a data loss closest to zero.

Of course, the previous example is entirely flexible, and it is always possible to have a cloud DR datacenter smaller than the one in production and only in case of contingency increase the DR server’s size to that of the production infrastructure. Recovery times are slightly higher but in return there are considerable savings in fixed costs required to maintain the DR infrastructure available, increasing only in case of a disaster.

Another factor to keep in mind is that public cloud services are usually distributed spanning different continents so that, with the appropriate considerations, it is possible to construct DR strategies preparing contingency datacenters with an extensive geographical distribution.
At Novis we have enough experience to guide your company when making this decision, based on the analysis of your business processes and current infrastructure.

As an example, one of the projects carried out by Novis was for a company in the Americas that sought to strengthen its DR solution. This client has two datacenters in one country and consequently has a disaster protection capability in a limited region. To complement and strengthen its DR strategy, a third datacenter with reduced capacities was set up on a public cloud in a geographical location distant from the disaster zone and at a reasonable cost, thus increasing the protection of their business processes from a catastrophe at country level.

Contact us, we can help you with your projects.
Feedback/discussion with the author: eduardo.morales@noviscorp.com, SAP Cloud Solutions Architect.