How To Transform Your Disaster Recovery Initiative With Hybrid Cloud

By Justin Mungal, AllCloud

Cloud-based solutions are driving the rapid growth of the disaster recovery as a service (DRaaS) market. In fact, this market has been growing at a compound annual growth rate of 44% and is expected to reach $23 billion by 2023.

Disaster recovery isn’t a new concept. As part of a business continuity strategy, all companies have some variety of DR in place to protect against hardware or software failure, a network or power outage, human error, or physical damage to a building caused by an environmental hazard like a fire or flood.

However, traditional on-premises DR requires in-house expertise and management, as well as infrastructure overhead that increases with data growth. Testing can be highly complex and disrupt the business.

Cloud DR is an exciting alternative because it offers numerous benefits over traditional DR including:

• Production infrastructure no longer needs to be duplicated and maintained in a secondary location, only a small Cloud environment with replicated storage
• You can test on-demand without impacting production environments
• Recovery can occur in minutes rather than hours or days, and the whole process can be automated.
• The consumption-based cost model of the Cloud means you only pay for what you use, without incurring large capital expenditures.

If you’re thinking about moving your DR to the cloud, it’s important to make sure you have the right building blocks to get there. Here are the top things you’ll want to consider.

1. Start By Assessing Your Business Requirements

First, the technology experts in your organization need to meet with your business experts to understand two key requirements. One is the Recovery Time Objective (RTO), which is the targeted amount of time you’re willing to take in recovering from a disaster. The second is the Recovery Point Objective (RPO), which refers to how much data loss you can tolerate in terms of time.

Traditional DR solutions typically target these objectives in a number of hours. For instance, if your RPO is six hours, your business leaders have said it’s acceptable to have six hours of data loss. But what if you could cut that loss down to mere minutes? That’s what’s so exciting about the recent capabilities of Cloud DR—and CloudEndure in particular—how it can take these recovery numbers down to a small fraction of what they used to be.

2. Create Your Testing Strategy

Testing is a critical factor that’s often neglected, especially in traditional DR environments, because it tends to be an expensive and time-intensive undertaking. And in traditional environments, you sometimes need to take your primary application down to perform the testing. Nevertheless, you could argue that testing is the most important consideration because without it you cannot be certain that your recovery processes will work successfully. And now, with Cloud DR, you can perform testing without impacting any production resources.

3. Consider Your Network Requirements

You also must evaluate what network requirements you have in going from the primary site to the DR site. Cloud solutions such as CloudEndure replicate changes in near real-time, which often use much less bandwidth than traditional DR solutions. That means you can meet strict recovery objectives for even your largest workloads, with less network overhead.

4. Evaluate Your Dependencies

Another important consideration is your server and application dependencies. In other words, you need to closely examine which of your servers depend on other servers, and which applications depend on other applications. Understanding and mapping out these relationships help you plan what order you need to bring up things in your DR site.

5. Look To The Cloud To See The Light

Cloud DR opens up a whole new world of possibilities to your business—one that doesn’t require replicating your entire infrastructure in the cloud. There are solutions you can evaluate that simply weren’t possible with traditional DR. One of those solutions is called Pilot Light, a design where you create a minimal infrastructure in the cloud, and you only provision the remaining infrastructure when you failover or perform a test. This type of solution is much more cost-effective as you only need to replicate your data changes. In addition, you can layer on automation that automates much of the testing and failover.

About The Author

Justin Mungal is an AWS Solutions Architect at AllCloud.