Disaster recovery as a service helps businesses guard against disruption
You’re sitting on an airplane parked at the gate and getting ready to go. Your seatbelt is fastened, your tray table is stowed, and your seatback is in the upright position for takeoff. But the plane isn’t moving. The forward door is still open, which isn’t a good sign. So what’s the hold-up?
Due to an IT system outage, the pilots cannot access the weight and balance calculations critical for their safety assessment. Without that data, no one is departing.
An actual outage of this system, used by many major airlines in the US, caused hundreds of flights to be delayed. The 40-minute outage had a ripple effect, causing a cascade of delays, missed connections all day, and extra work for airline employees. The provider itself stated that “just five minutes of system downtime can result in over 100 delayed flights and loss of revenue.”1
What that IT system lacked was effective disaster recovery (DR). We all recognize the importance of backing up our data, but businesses can suffer severe financial losses when they can’t get back up and running quickly. A single hour of server downtime costs about $300,000 for a mid-sized enterprise. But hourly outage costs can exceed $1M for large enterprises.2 That also doesn’t account for less-tangible harm, like damage to a business’ reputation. If data is lost because of an outage, that can also lead to problems ranging from missing transactions to regulatory fines.
Despite our best efforts, outages happen. Whether it’s a hurricane, power outage, cyberattack, or a glitch in the software, you need to be prepared.
Because of the high cost of downtime, most organizations (hopefully all!) have some system for backups, disaster recovery, and business continuity. The 3-2-1 backup rule is common guidance that prescribes:
- Three copies of your data—one production and two backups
- Two different backup formats, such as a local hard drive and in the cloud
- One backup stored offsite
This is a sensible plan, but it can be difficult for a business to manage. In the early days of disaster recovery, backups were often hand-delivered to another location. In today’s digital world, backups must be updated much more frequently, leading to businesses maintaining their own backup systems.
Now, disaster recovery as a service (DRaaS) is becoming more common. These third-party service providers manage the storage and backup tools, originally to their own data centers and now frequently in hyperscale clouds.
The global DRaaS market is growing rapidly, expected to scale from $8.8 billion in 2022 to $23.5 billion by 2027.3 According to analysts, growth is fueled by a desire to lower total cost of ownership (TCO), save time, and focus IT resources on higher-value tasks.
DRaaS vs. DR
Traditional disaster recovery is frequently self-hosted, meaning an organization must mimic its existing production environment. In fact, following the 3-2-1 rule, they need to replicate their production environment twice! The disaster recovery system needs IT support staff, administration, facilities, and infrastructure, and the solution can get expensive quickly as hardware, software, real estate, utilities, and personnel costs add up.
While a DRaaS provider also has these costs, they can spread them across their customers to offer economies of scale. Using multi-tenant cloud architecture can further reduce the need for equipment, space, and personnel to lower costs.
DRaaS also differs from DR by offering on-demand Infrastructure as a Service (IaaS), recovery service level agreements (SLAs), and automated recovery and fallback.4 A cloud-based DRaaS that includes IaaS will be able to replicate network infrastructure as well as databases and applications, making it more reliable than traditional DR.5 Recovery SLAs from a DRaaS provider guarantee you’ll be back up and running within a specified time frame.
Two key metrics for DRaaS SLAs are recovery time objective (RTO) and recovery point objective (RPO). RTO is the amount of time before you’ll be up and running again in the event of a disaster. RPO is essentially the frequency of backups. For example, if a business has a major outage at noon with an RTO of two hours and an RPO of one hour, it should be back up and running by 2pm based on data captured at 11am or later.
The chosen times for RTO and RPO are highly dependent upon the type of data or service involved, as well as what the business can afford. A busy bank would need a very short RPO so transactions wouldn’t be lost, but an infrequently-updated website could have an RPO of weeks. In the case of the aircraft weight and balance calculation system, a very short RTO would have been more beneficial than a short RPO, as history is less important than availability for that system.
While every business needs a disaster recovery solution, many factors, from cost to ease of use, should be considered when making a selection. With the right solution in place, you’ll be prepared for anything that comes your way.
- Conde Nast Traveler, This Latest Outage Shows How Little We Know About Air Travel, April 2019
- ITIC, Hourly cost of downtime, 2021
- MarketsandMarkets, DRaaS Market worth $23.5 billion by 2027, July 2022
- Gartner, Why You Need To Know About The Disaster-Recovery-as-a-Service World, June 2020
- Security Magazine, Disaster Recovery as a Service: What is it and do you need it?, March 2021