Disaster Recovery: Ditch Your Secondary Data Center for Cloud

Laz Vekiarides is Co-founder and Chief Technology Officer at ClearSky Data.

When the concept of disaster recovery (DR) first came about, there was really only one way to accomplish it: Build it yourself. When the corporate data center was the only place a company's critical data could live -- and the only place its most important applications could run -- keeping that data center up and running with five-nines reliability was worth any effort and almost any price. Doing it right often included building a completely separate second data center for data protection to house all the redundant copies of data, all offsite backup, and complete DR systems.

Building a secondary data center just for DR is no longer a sustainable option, nor is it a necessary one. Thanks to the growing cloud adoption rate and cloud-forward co-lo providers, the data center world is transforming, which means more efficient and cost-effective options, such as hybrid DR.

That’s not to say that moving from an on-premises secondary data center to the cloud or a hybrid model happens with a snap of the fingers. It can be a complex process that has many moving parts, and the consequences of mistakes are severe. So, how do you make migrating your DR to the cloud as uneventful as possible? Having a plan is the surest way to ensure migration success.

Before you make a single change to your DR structure, break down and evaluate every aspect of your IT strategy, starting with immediate needs and continuing on to resource requirements and long-term goals. Here are some things to consider in your plan, before, during and after you move to the cloud.

Evaluate Cloud Benefits

The first thing to look at is what exactly you’ll gain from getting rid of that secondary data center and moving your DR to the cloud. If you're going from a secondary data center that you’ve been managing for years to a more modern, hybrid environment, the benefits likely start with a simpler infrastructure and access plan. There are also benefits to your actual ability to recover from a disaster, including data that’s backed up to the cloud and is stored in virtual machines making access much easier.

On the cost side, many organizations can save 50 percent -- or more -- by moving to a disaster recovery as a service (DRaaS) option in the cloud. But, like most things in IT, it’s not that simple. There are a few specific factors, in particular, to look at:

Software license and support agreements: These agreements can be lengthy and complicated, and may have special considerations that affect a potential move to the cloud. It’s important, then, to review the cost of everything license related: replication licenses, backup and DR software licenses, and even WAN and other technology costs. Consider what you can get rid of when you move to the cloud for DR.

Personnel skills and options: One of the benefits of a DRaaS option is that it’s easier to manage. However, with any new technology, there’s a ramp-up process. Options like a managed service provider with expertise in DR can help you create and secure your new environment while your staff gets acclimated to the new way of doing DR.

Your company-specific costs: Besides the factors mentioned previously, it’s important to consider any costs that are specific to your company. Maybe you have a long-term lease that you would need to break. Maybe there are other business impacts of shuttering your secondary site. Make sure you know all your obligations before you move, so you can fully measure the costs and benefits in your analysis.

Consider a Hybrid Cloud Approach

One of the many benefits of moving out of your secondary data center and into a hybrid cloud situation is the opportunity to improve your IT performance through a multi-cloud approach. Working with different cloud providers for different aspects of your business is a great way to reduce costs and avoid relying on any single vendor too much, which has the added benefit of reducing the chances of a catastrophe should one vendor have an outage.

It’s also important to remember the benefits of telecom carrier diversity when considering cloud provider options. Carrier competition -- even within colocation sites -- can help keep costs low and ensure you’re getting the best possible features and services available.

Develop Objectives to Measure DR Success

The final piece of a good plan is how you will measure success. With DR, it comes down to a couple of essential metrics:

●      RTO - How long can your critical IT resources be down before it affects your company in a significant way?

●      RPO - How much data loss can your organization afford in case of a disaster?

For most organizations, the answer for the time question is under a minute. For data loss, the answer is likely similar. Is it OK for your business if you’re restoring data from a backup that’s hours old? For most applications, the answer is, “No.”

When you’re planning to move your DR to the cloud, then, make sure the provider you choose can help you meet these objectives. This is something else that can be helped by a hybrid approach. For example, making use of edge computing options -- where critical data is stored at the edge of the cloud, close to your physical location -- can increase your disaster readiness, and allow you to recover that critical data without missing a beat.

DR has long been the bane of many IT departments’ existence. You hope never to use it, but you spend significant time and resources building, managing, testing and maintaining secondary sites, which in the end may not even help you recover from an outage quickly enough.

Just as it has for many other IT issues, the cloud has stepped in to provide a better way to handle DR. As-a-service and hybrid models can help organizations save significant time, money and resources on DR, while at the same time improving their overall disaster preparedness. So, never build or manage a secondary data center again. For true disaster readiness, head to the hybrid cloud.

Opinions expressed in the article above do not necessarily reflect the opinions of Data Center Knowledge and Informa.

Industry Perspectives is a content channel at Data Center Knowledge highlighting thought leadership in the data center arena. See our guidelines and submission process for information on participating.