As disaster recovery as a service (DRaaS) becomes more accepted, a wider variety of organizations are starting to see its benefits and consider it for their backup and disaster recovery needs. There’s no doubt it offers significant cost and aggravation savings over traditional methods, not to mention it’s better at actually helping you recover from an outage.
With so many considering it, however, there can often be a rush to implement a solution without looking carefully at the factors that will help you make the best choices on what type of DRaaS works best for you.
Recently, George Crump of Storage Switzerland tackled this issue in article about capabilities to look for in a DRaaS offering. He provides some real insight into DRaaS as a whole, including these three areas to look for in DRaaS and how to use them:
1) Complete platform protection. “When going with a DRaaS provider, it's important that the DRaaS approach either complement or replace the existing backup tool.”
This is the whole point of disaster recovery as a service, right? Protecting yourself if something goes wrong. It’s really an insurance policy for when something does go wrong. Even in a 100 percent cloud environment, something will eventually happen; even cloud providers will have an outage.
The best way to prepare for this is to have as much of your infrastructure and software as possible on different providers. For example, if you’re 100 percent on Amazon in US1 West, consider making your DRaaS Azure Midwest. Likewise, if you have on-premise infrastructure and your primary is in New York City, consider San Francisco for colocation. It’s the same concept; the idea behind DRaaS is that some piece of hardware will inevitably break. It always does. This doesn’t change whether you're all cloud, all on-premise or hybrid. Even the cloud is infrastructure, just run by someone else.
There’s another benefit, too: working with multiple vendors can increase performance, and it lets you avoid vendor lock-in. Put another way, today, hybrid cloud really means multi-cloud. This applies to carrier diversity, as well. When carriers are competing for your business, even in a single colocation site, you’re more likely to get the best features and prices.
2) Recovery time objective and recovery point objective. “RTO and RPO promises range wildly among DRaaS providers. IT must examine vendors' processes to understand what they can reasonably deliver.”
The first step after a disaster happens is getting back online, right? This is another area where DRaaS can potentially shine, if you make the right decisions before disaster strikes. Certainly, as Crump says, it’s critical to completely understand what a vendor can realistically offer when it comes to metrics like RPO and RTO.
Let’s look at it from the other side, though. Customers must know what they need for both metrics before going into a DRaaS solution. This means things like defining how often images are backed up and what’s acceptable downtime. If a DRaaS system is taking an image of your entire system every 15 minutes, your RPO is never any more than that, and is likely less.
RTO depends on things like where your backup data is stored: in inexpensive, but slow and distant facilities, or closer and on higher performing media. Likely, what you want is a combination where your infrequently used (cold) data is stored in the most cost-effective way, and your frequently accessed data (warm and hot) is closer and ready to be used at a moment’s notice.
Determining what combination works best for you, while keeping your RPO and RTO at manageable levels requires some work on the front end, but it’s worth it when an outage occurs. This will vary widely from company to company, so make sure you know what you need, then make sure any provider you choose can keep up.
3) Return to normal. “Another area to consider is how the DRaaS offering will handle a return to normal operations.”
Above and beyond RTO and RPO, there is the need to return to normal after an outage. As Crump mentions in his article, a great benefit of DRaaS is being able to run in the cloud indefinitely, for example if a data center is hit with a major incident and knocked out. If you are continuing in the cloud for any length of time, issues like latency can come into play. Edge computing and a hybrid data approach – like we mentioned previously for cold, warm and hot data – can help you overcome many issues, while hanging on to both disaster recovery cost savings and performance metrics.
Depending on your data footprint, the effort to get back to normal can be staggering. The transfer of backup data and a so-called fast-sync of things that have changed since the outage can take weeks. If your company has complex data and application needs, it’s important to consider this fact when deciding on a DRaaS solution. Features like bulk shipments of your original dataset can cut down restore time significantly, getting you back to your normal operations that much quicker.
These three considerations should provide some areas to think about, whether you’re just getting started with DRaaS or you’re giving your system a second look. Wherever you are in the process, remember there’s no shortcut to sitting down and determining the functionalities you need, in addition to what the vendor provides.
To learn more about what DRaaS features are right for your company, contact us.