I recently had a chance to talk with Rick Vanover of Veeam Software about what businesses need to do these days to ensure their availability strategy fully addresses their needs. Rick is the director of product strategy at Veeam, where he leads a team of technologists and analysts that brings Veeam solutions to market and works with customers, partners, and R&D teams around the world. You can follow Rick on Twitter @RickVanover.
MITCH: Rick, thanks very much for agreeing to let me interview you on the topic of how data protection and availability are changing in our cloud-based area and what organizations are doing right and wrong these days when it comes to preparing for disaster recovery.
RICK: My pleasure. This is an area that I’ve built my career around and organizations today need to ensure that their availability strategy meets the expectations of the business.
MITCH: The last few years have seen a lot of changes in how organizations of all sizes implement and manage their IT infrastructures. Cloud computing models like software as a service (SaaS) now enables users to connect to and use cloud-based apps directly over the Internet with Microsoft’s Office 365 being one popular solution of this type. Then there’s infrastructure as a service (IaaS) that lets organizations build agile computing infrastructures that can scale up and down on demand. What does and doesn’t change in regards to ensuring data protection and availability for your business with these new models?
RICK: This is a great question, Mitch, and I’m glad you asked it. The one important thing I’ve learned over the years is that while the platform may change, the rules on the data and availability expectations do not change. Microsoft Office 365 is a good example that you have illustrated. The promise of this Software as a Service (SaaS) solution is great: a great relief of on-premises tier 1 storage, the opportunity to reduce the need for mailbox quotas and with OneDrive for Business a way to combat “shadow IT” file sharing outside of corporate mechanisms. These are real business problems solved by Microsoft Office 365 and these changes are welcome to both users and IT administrators alike.
But what doesn’t change when the application does change? The responsibility of the data. Organizations need to realize that this is their data and Veeam has invested in a new product, Veeam Backup for Microsoft Office 365.
MITCH: What sort of changes do organizations need to make in their supporting processes to ensure data protection/availability in the event of a disaster when they’ve embraced the cloud wholeheartedly or at least adopted some sort of hybrid IT model?
RICK: As the mix of platforms change for organizations, the disaster recovery aspect absolutely needs to be reassessed. This is a very difficult task and, honestly, the old way of doing this isn’t acceptable anymore today. I know plenty of IT administrators who addressed disaster recovery as a once-a-year test where there was free pizza over the weekend, things were tested, about half of it failed, and the goal was to do better next year. Today’s IT services and expectations can’t deal with that.
I know plenty of IT administrators who addressed disaster recovery as a once-a-year test where there was free pizza over the weekend, things were tested, about half of it failed, and the goal was to do better next year. Today’s IT services and expectations can’t deal with that.
This is one reason Veeam have developed a new product that went available earlier this year, Veeam Availability Orchestrator. This product brings a very critical capability for disaster recovery in the era of hybrid IT. Veeam Availability Orchestrator supports orchestrating disaster recovery for on-premises workloads; but also supports orchestrated disaster recovery to VMware Cloud on AWS. This is a new cloud offering in Amazon for VMware workloads. This is an example where an organization can have their on-premises resources benefit from DR in the cloud — literally!
MITCH: I’ve heard it said by some who provide IT support for businesses that rely mostly on SaaS applications that “backup” is basically a bad word now, that performing daily backups is a dead practice because there are now more sophisticated ways to ensure availability for your business data. But this sounds a lot like an oversimplification to me. Does the availability burden of performing regular backups really go away with cloud computing?
Backup and recovery, as well as replication and failover, are the important critical functions there. This effectively is a gateway to more advanced capabilities.
The next step is an aggregation of those critical data sources, whether they are in the cloud, on-premises, or in the SaaS space. Having the data flow for all critical data is an important milestone, and each platform has their own characteristics that may change what the capabilities are for backup and recovery.
With the aggregation of this data, visibility becomes important. Answering key questions like what data is where, who is accessing what, will the environment run out of storage and such are very critical questions today.
Advanced capabilities become the next opportunity, and orchestration is a capability that Veeam brings today to the market that can respond to changes very easily. For example, orchestrated disaster recovery to another site can be done with confidence if there is a concern that weather is going to take out a data center, so organizations can proactively fail overconfidently.
With all of these capabilities, then automation becomes the goal. Automatic resolution of issues and policy violations, for example, will be a capability from Veeam later this year. This can be very important when it comes to ensuring that critical data is protected to the level the business demands today.
This is a quick overview of our vision, but it is important to reinforce that it all starts with a solid backup and recovery as well as replication and failover capability.
MITCH: Given all these changes that are happening, what are most organizations doing right and wrong these days with regard to disaster recovery?
RICK: My observation today is that many organizations are simply not providing the availability experience their business demands. The best example is a high-speed recovery technique. Ask this question: If someone accidentally deletes a virtual machine, how soon can it come back? If the answer is more than a few minutes, there is a gap between capabilities of what is in place and the expectation of users. That’s one example that Veeam has pioneered and led the market for over eight years and there are move. Same for an AWS EC2 instance in the cloud: If someone terminates and deletes it, how soon can it come back? If that too is more than minutes, there is a gap.
Organizations are doing things right when it comes to leveraging new platforms. This includes leveraging SaaS applications where it makes sense, leveraging the cloud and leveraging service providers for the right services as well.
The key advice I have to offer is to ensure that availability is thought of every step of the way.
MITCH: How do new platforms like hyper-converged infrastructure (HCI), advanced storage systems, and robust networking change the game in regards to availability?
RICK: They provide new “plumbing” to work with. New APIs, new networking techniques, and better snapshot capabilities. These are all important for efficient data flow for backup and recovery as well as replication and failover.
Veeam has always invested in APIs and platform capabilities to address data movement at scale. The new technology platforms in the mix are built in the same mindset as well.
MITCH: Looking ahead then, what do you see in the future for business continuity and disaster recovery?
RICK: I see a continued push for completeness. Organizations will make changes to applications, data, and other critical systems to make them more “DR friendly.” A good example is a legacy application that sits on a physical server that is obsolete on an operating system out of support for three years. Can that application really have good DR? No.
The key advice I have to offer is to ensure that availability is thought of every step of the way.
Organizations are indeed seeing the value of proper DR and if the application needs modernized, changed, or sunset out of production use, that’s what it takes.
Proper DR comes with modern platforms and data; obsolete components can’t be made awesome!
MITCH: What practical advice would you give to an admin for implementing disaster recovery solutions in a hybrid cloud environment? Any tips or recommendations to help them get it right going forward?
RICK: If there is a gap in the availability strategy, my advice is to start small and make it right. Specifically, that would mean take one small application. Get the basics of backup and recovery right. Then set up replication and DR capabilities for that small application. Once you have that working right, that motion educates organizations about what solid backup looks like. How those types of tools work, etc.
Then move to the next application that is a bit more complex. And succeed. Get applications to the right model one at a time. Don’t start with the biggest, most critical application in the mix at first.
Once the small successes are proven, go back to the business (like operational people in an organization) and indicate that proper DR and better backup and restore times can be achieved if we virtualize this application or invest in a storage snapshot engine or such.
If the business is drawn to the benefits, the effort to change the platform may come much easier.
MITCH: Rick, thanks very much for giving us some of your valuable time!
RICK: Cheers, Mitch, my pleasure.