Companies who require high levels of availability need a reliable disaster recovery strategy. For many organisations, that usually means adhering to the 3-2-1 backup rule, where the 1 refers to keeping 1 backup copy offsite. Traditionally, that copy is stored in a dedicated, company-owned and/or managed disaster recovery infrastructure. Lately, however, businesses have started turning to cloud-based disaster recovery solutions. This article explains why that strategy is better.

What is the 3-2-1 backup rule?

The 3-2-1 backup rule is a highly recommended backup strategy that says you must:

• Possess at least 3 copies of your data;
• Store copies on two different kinds of storage media; and
• Keep one of those copies offsite.

The rationale behind having 3 or more copies of your data is simple. The more backup copies you have, the less likely you’ll lose all copies to theft, misplacement, a natural calamity, a cyber-attack, or any unfortunate incident.

Why store copies in at least 2 different kinds of media? Well, if they’re stored on backup solutions based on the same media types, the two copies will be subject to the same physical characteristics, limitations, and vulnerabilities of that medium.

Lastly, it’s advisable to store at least one copy at a different geographical location because, if the site where the primary copy is stored becomes inundated (e.g. submerged in flood waters), that offsite copy will still be safe and usable.

Disadvantages of building your own offsite DR facility

In keeping with the 3-2-1 rule (particularly the ‘1’ part of the rule), enterprises typically build their own disaster recovery (DR) site and keep copies of their backups there. Depending on the type of DR site (hot, warm, or cold), the entire facility may include servers and networking equipment, as well as provisions for power and HVAC air conditioning. These elements plus the building itself will surely entail a huge capital investment.

The large CAPEX requirement isn’t the only thing you need to worry about when building your own DR facility. Once that facility is up and running, you will have to allocate OPEX for it. The DR site’s operating expenses will, at the minimum, include costs for power and cooling, as well as salaries for the IT staff assigned to manage that site. 

Advantages of using cloud-based disaster recovery

Low upfront costs

The most obvious benefit of pursuing a cloud-based disaster recovery strategy is that it will allow you to save on upfront costs. Cloud storage requires very minimal or even zero upfront costs because it follows a pay-as-you-go pricing model.

Unlike in a traditional DR facility, where you will have to build an infrastructure with almost the same capacity as your main site, a cloud-based DR solution will mainly require only replicated data and a substantially scaled-down version of your main site consisting of a few virtual machines. The only time you will need to spin up more virtual machines (and pay for them) is when you do DR testing or if an actual disaster recovery operation is underway.

Lower operating expenses, greater value

Because a cloud-based disaster recovery infrastructure will likely be shared by several customers, the Disaster Recovery as a Service (DRaaS) provider can take advantage of economies of scale. As a result, the costs for storage, Internet bandwidth, cooling, staff salaries, and others, that are passed on to customers, are minimal.

The ability to take advantage of economies of scale also allows some cloud backup and disaster recovery providers to offer additional functions at no extra cost. For example, in Serviced Cloud’s backup and disaster recovery offering, data is backed up in 4 places through what is known as quadruplication. In addition, backups and replicas are transmitted over the WAN through WAN acceleration, which allows customers to overcome bandwidth limitations.

Affordable DR testing

Testing is a crucial component of any disaster recovery program. Through testing, you will be able to identify possible flaws in your DR plan and, in turn, rectify them. Testing will enable you to pinpoint areas for improvement.

There are different kinds of DR testing methods. There are methods that simply entail reviews and walkthroughs of the DR plan. However, there are also methods that require actual testing of all hardware, software, and personnel involved in the disaster recovery process.

A full simulation, for example, may require participation of the entire DR facility. In a traditional DR strategy, that DR facility must already exist. But in a cloud-based DR, only a few virtual machines may be provisioned initially. And then during the simulation (or immediately prior), a full-scale replica of the “DR facility” may be constructed by simply spinning up more VMs. After the simulation, those VMs can then be spinned down. 

During the simulation, you will have to pay for the cost of those additional VMs. But after the simulation and throughout the rest of the year, those VMs will be spinned down and won’t incur any costs whatsoever.