DRaaS advice – Cloud-based disaster recovery

First, the Basics

As with anything that involves marketing and vendor solutions, different definitions of DRaaS exist. PC Magazine describes it as, "an online service that specialises in setting up a company's applications in the event of a catastrophe."

Similar to backup, many DRaaS solutions involve placing remote copies of an organisation's data and systems at a service provider's cloud data center or colocation site. Yet this is where the similarity to backup usually ends. According to an article in TechTarget, 75% of backup vendors' recovery efforts involve recovery of individual files from the past 30 days. In contrast, the practice of disaster recovery is about how well (and how fast) a company's key systems and applications can be brought back online after they experience a significant outage or disaster.

Many technological variations of DRaaS are available. There are self-managed methods where a cloud vendor provides the IT infrastructure to store a customer's DR files but leaves most of the subsequent disaster recovery heavy lifting to the customer organisation. There are also offerings that can replicate physical servers, virtual servers or virtual machines (VMs) from a primary data centre to a cloud provider. The provider can then "spin up" the VMs in the event of disaster — even offering IPSec tunneling or SSL VPN connections so users can access the remote VMs.

Then there's a higher level of managed DRaaS offerings, which might involve the above features and more. For example, service providers might offer more flexible DRaaS platforms to meet each client's unique DR needs. They may have pre- and post-DR planning services, detailed client data onboarding, detailed DR process runbooks, DR testing, detailed declaration procedures and guaranteed recovery times for critical systems.

Nail Down What You Need from DR

A good piece of DRaaS advice for organisations trying to evaluate the merits of one provider over another is to first fully understand what you need for disaster recovery and business continuity. It's also wise to review the recovery time objective (RTO) and recovery point objective (RPO) for different application tiers. Also, how will virtual machines factor into recovery? Be sure to understand the meaning of terms like replication, hot site, warm site, failover, failback, and mirroring.

What to Look for in a DRaaS Provider

As every enterprise is different, your situation will have unique needs. However, the following list is a good starting point for evaluating whether or not a provider will make the cut.

• Credibility: Look for trusted, proven DRaaS providers your organisation feels confident relying on for recovering its critical systems and data.
• Flexible, "Best-of-Breed" DRaaS Technology: Your organisation needs to know that its service provider's technology is proven to be effective and will work for your environment. You also need to know that it can accomplish exactly what's required.
• Security: It's important to be confident in the levels of physical and logical security your service provider uses to keep your data safe, protected, encrypted and immune from theft and most disasters.
• SLA Guarantees: When the rubber meets the road, a good DRaaS provider will spell it all out in detail within the customer service level agreement (SLA). This includes any time-based guarantees for recovery of systems after an official disaster is declared.
• Affordability: A good DRaaS provider will make creative efforts to meet each organisation's DR budget. Service providers should also factor in data growth rates and offer ways to grow DRaaS services to offer organisations a good ROI. That being said, it's important to note that many analysts recommend not letting cost be the primary factor for DRaaS provider decisions.

While judging providers along these lines will certainly point you in the right direction, you may want to go deeper. This webcast may offer additional insight as you prepare to take this important step in your organisation's disaster recovery planning.