Why is Disaster Recovery Planning Important?
Today more than ever, organizations are choosing to move business critical IT workloads from traditional on-premises data centers to the public cloud, and adopting hybrid cloud models to operate business critical functions.
While the public cloud can provide a great deal more flexibility than a traditional data center, it is important to understand that disaster recovery capability is not built-in to the services offered by cloud providers like Amazon Web Services. Instead, when building in the cloud, application teams must explicitly implement a DR strategy themselves. Thus, a move to a hybrid cloud model introduces additional complexity to DR planning, as the risks associated with both on-prem and cloud environments must be considered. These differences need to be understood and addressed proactively and, in some cases, independently in order to ensure the ability to recover from a disaster in any part of your hybrid cloud environment.
What Kinds of IT Disasters Should I Plan For?
IT disasters come in 2 forms: infrastructure outages and cyber attacks. Your disaster recovery plan should account for both, across both your cloud and on-prem infrastructure.
The first type of disaster to plan for is an infrastructure outage. This could be a traditional on-premises incident, such as a data center fire. But an infrastructure outage can also occur in the public cloud, in the form of unexpected downtime of your cloud provider. In AWS, for example, major outages have happened in recent years that have cascaded across shared services and availability zones.
Regardless of the location of the infrastructure outage, to recover from these disasters, you need the ability to fail over to alternate infrastructure that has no shared dependencies with your primary infrastructure.
The second kind of disaster comes from a cyber attack – any time a bad actor gets into your IT environment, like ransomware. To recover from these disasters, whether on-prem or in the cloud, you will need the ability to restore your historical data into a clean environment that hasn’t also been compromised. You don’t want to restore to the same infrastructure – that environment has been contaminated.
Building Disaster Resilience
So how do you prepare for both of these scenarios in your hybrid cloud environment?
Your on-premise and cloud-hosted workloads run on independent infrastructure, so in a hybrid cloud scenario, you solve for infrastructure disasters independently. If your datacenter goes down, you must fail over to a different datacenter. You can also fail over to a recovery environment that is provisioned in the cloud. If your public cloud infrastructure goes down, you will fail over to a different part of the cloud, which is usually in an alternate region to hedge against the regional failure scenario outlined above.
In preparing for a cyber disaster, however, you have to assume that the inter-networked environments in your hybrid cloud will be compromised together, and you’ll need to have a plan to recover them both to uncontaminated infrastructure.
Watch below as CEO Doug Neumann breaks down how to solve for different disaster scenarios in the hybrid cloud.
Arpio’s Solution Makes DR for Your Amazon-hosted Infrastructure Easy
Many organizations feel daunted by disaster recovery planning for a hybrid cloud model. While DR for on-premises workloads is by now a well-understood concept, the introduction of cloud-native tools that make the public cloud so flexible can also make it an infinitely more complex environment to protect. The ease of creating applications in the cloud also introduces new governance issues in larger organizations who are trying to ensure a recovery plan. And, if application teams have not built their architecture to account for all of the disaster scenarios from the beginning, it can be costly and time consuming to develop and retrofit.
While there is no silver bullet to solve for all of the challenges that come with DR planning for the hybrid cloud, Arpio was designed to make DR for AWS’s public cloud easier. Arpio has built the most comprehensive Disaster Recovery solution on the market. Simply connect your AWS account to Arpio using our validated security model, and Arpio can do the rest. Our software will quickly inventory your entire infrastructure and generate a list of resources for protection. Click what’s important to you, and Arpio will discover all of their dependencies and immediately begin curating your recovery environment for you. No engineering necessary.
A Single Solution for Hybrid Clouds Using AWS
Because of the differences in outage and recovery types across on-premises and cloud architectures, it can be difficult to envision a single solution that can work across your hybrid footprint.
However, it is possible to implement a single disaster recovery solution to protect your hybrid environment from both types of disasters – and Arpio can help.
In general, a hybrid cloud relies on hub-and-spoke interconnectivity between your on-premises workloads and those that run in the public cloud. One example of this kind of service is an AWS Transit Gateway.
To build out your Disaster Recovery plan around this kind of hub-and-spoke model, you can use Arpio to replicate the “spokes” into recovery environments that are running on independent infrastructure, in alternate cloud environments. This model means that your cloud environment will be insulated from the disaster scenarios we laid out above.
In the case of a disaster, you won’t interconnect these replicated spokes until and unless they’re needed. When a disaster strikes, and you determine whether you’ll be performing a partial failover or a complete failover, you can use Arpio’s push-button automation to launch the recovery spoke environments and reconfigure the inter-networking just in time.