Disaster recovery has come to mean one of two things:
Either you build and maintain your own separate data centre, logically and physically separated so you can manage and execute a fail over in case of disaster, or you leverage a third-party service provider’s recovery cloud for your data protection and recovery.
Both have their pros and cons and make sense for different organisations. But a third option has emerged that has caught a lot of attention: recovery to the public cloud.
Why are organisations turning to the public cloud? And how is it transforming disaster recovery? Here’s a quick guide to what’s creating so much buzz.
What is recovery to public cloud?
The public or hyperscale cloud -- AWS, Microsoft Azure and Google Cloud, for example -- has become a destination for certain workloads. Public cloud’s pay-as-you-go pricing makes it accessible, especially for organisations still growing their infrastructure or that need to scale their server usage up or down during peak business.
There’s no upfront hardware capital investment, which saves money and time. Larger companies have been using the public cloud to test their development environments and increasingly for production workloads, even if some of them have to keep more sensitive data in their private cloud for compliance-related reasons.
Disaster recovery to the public cloud can leverage the usage-based economics while achieving aggressive recovery time objectives (RTOs) and near-real-time recovery point objectives (RPOs). It’s an attractive model that’s been gaining momentum especially in the last year.
Even for organisations hesitant to put workloads in the public cloud, leveraging the public cloud for recovery first gives them a way to test the waters with their disaster recovery assets before moving production workloads.
The case for recovering to public cloud
With more and more companies looking to public cloud as part of their disaster recovery strategy, it’s important to understand why -- and how your organisation might benefit as well. Here are five reasons companies are eager to leverage the benefits of public cloud for DR.
- Usage-based economics. The capital expense of running your own data centre has been well documented. By using the hyperscale cloud, you can free up time and money that would otherwise be funneled into maintaining and expanding your own systems. Using the public cloud, compute resources are not charged until At Time of Test (ATOT) or At Time of Disaster (ATOD), which means you are not paying for idle resources that you are not using.
- Fast recovery times. There are automated orchestration technologies that enable recovery in AWS cloud environments in as little as 15 minutes, helping you meet your overall RTO. With the cost of downtime estimated at $336,000 per hour, every minute counts.
- Reliable data replication and backups. There are technologies that provide Continuous Data Protection (CDP) and replication of data, providing near real-time RPOs that guard against data loss. In addition, these technologies also allow organisations to retain multiple snapshots of data -- for example, snapshots every 10 minutes, hourly and daily over a rolling data retention period (e.g., 30 days) -- to guard as much critical data as possible should a cyberattack or natural disaster compromise your data centre.
- Compliance support. PCI DSS. HIPAA. Recovery to the public cloud doesn’t mean risking compliance violations. The security controls that public cloud providers have in place enable you to achieve compliance possibly even more easily than in your own data centre.
- Hybrid recovery. Just because public cloud is an option doesn’t mean everything should go there. Just as organisations run applications both in the cloud and on-premise, some environments might be recovered to an organisation’s on-premise -- or service provider -- data centre and others in the public cloud. You just have to make sure that your applications get back up and running with components potentially recovering to different locations.
How to get started with recovery to public cloud
If your organisation lacks recovery or public cloud skills, a managed solution is the best option for immediate improved resiliency. That way your source data is always protected with a fully managed, automated and service-level agreement (SLA)-backed replication and recovery solution.
The process begins with discovery to determine and validate company requirements and start standing up the necessary cloud infrastructure. The build phase includes software installation and configuration for your servers that need to be protected, as well as configuration of the recovery plan and settings before starting replication.
Once replication is complete, initial testing occurs and RTO is benchmarked. At that point, the replication is managed, monitored and maintained while the architecture is continuously improved. Regular testing identifies any issues and quickly mitigates them, and the system is ready to respond swiftly in the event of a disaster.
Recovering to public cloud is gaining traction, both for its economics and its speed of recovery, which are changing disaster recovery as we know it. If you’re ready to get started, learn more about Cloud Recovery -- AWS from Sungard AS.