An overview of fail over models for data and visualized infrastructure, plus a discussion of SLAs and tiered recovery.
There's no one reason why companies move to the cloud. You hear many reasons: cost is often mentioned, although a bit less these days. You often hear security cited, as providers have more to spend on technical excellence. And you certainly hear flexibility mentioned; it's probably the main reason at present.
What's not generally spoken about is the ease of disaster recovery, yet this is something that touches on all the elements mentioned above. By replicating virtual machines into a private cloud, you're reducing costs, certainly compared with the cost of installing physical hardware. It's a flexible solution, as it can scale to cope with unexpected demands. And finally, by using a variety of security tools on the network, you can ensure that your data is safe from prying eyes.
Many businesses have discovered these advantages. According to the ZDNet cloud priorities survey, a third of organisations said backup and recovery was the most important infrastructure as a service (IaaS) function to them; it was the leading choice.
But that's not the whole story; cloud on its own doesn't cut it. Smart companies use a hybrid approach to maximise performance and minimise costs. In this scenario, mission-critical data is backed up on-premise and then, automatically, transferred to the cloud over time. A hybrid product like Microsoft's StorSimple can support this approach, transferring data to the most appropriate medium. In this case, the decision-making is taken away from the CIO, and the solution is both cost-effective and efficient. It also means there is no need to calculate how much capacity is needed - that in itself gets rid of a major headache for the CIO.
Recent research supports the rapid uptake of cloud backup. An April 2015 survey by Forrester found that 56 percent of companies were backing up tier-2 applications at least once a day, a rate that has doubled in the past three years.
In the end it's the hybrid approach that is going to bring in dividends: it's not a case of all public or all private but the best fit. Forrester points out that companies with large databases which require frequent backups should stick with on-premise solutions while medium-sized tier-2, tier-2, and tier-3 datasets could all be moved to the public cloud.