Disaster recovery as a service (DRaaS) continues to grow on both the demand and the supply side of the equation. While some still wonder if DRaaS makes sense for their business, many customers and vendors alike are flocking to the space. I personally feel that DRaaS makes perfect sense for many companies for a lot of reasons, so let me explain some of them here.
The first reason that DRaaS makes sense is that it’s possible due to virtualization. It wasn’t that long ago that “DRaaS” (we didn’t call it that back then) meant a vendor needed an army of different kinds of physical servers to be able to serve your DR needs. Virtualization simplified the entire setup if for no other reason than most people using virtualization standardized on Intel, Windows, and Linux. That’s so much easier to provide as a service than an array of servers capable of running Solaris, AIX, HP-UX, etc.
As long as your DRaaS company supports the same hypervisor you use, they will have the exact same virtual hardware configuration you have in a DR scenario. It is also possible that they can convert your VMs to their hypervisor’s format. If you do, there will be a slight change in the hardware your OS thinks it’s running on, as each hypervisor has a slightly different virtual hardware configuration they present to the VM. It’s easier and cheaper for them to support you, and it’s easier for you to use them. (Not everyone has gone to Intel, though, and any DRaaS vendor needs to consider that. More on this later.)
DRaaS is also possible due to a number of advances in backup and recovery technology, including byte-level deduplication and bandwidth optimization. Even more important is the idea of storing the backup in a format that allows it to be immediately usable in a disaster. So called live recovery or instant recovery is a major advancement in backup and DR technology.
Historically, backup software only knew how to store data in an encapsulated backup format such as TAR. This meant that in an outage, we first had to repair or replace the hardware that was damaged in the disaster, and then we could begin the process of restoring the data. Both processes took many hours, even days, to complete. With a combination of instant recovery and virtualization, the original systems don’t need to be repaired or replaced. You simply point a virtualization engine at the backup copy of your data and run the VMs from there. You can recover from a small disaster in minutes and a major disaster in a few hours by running everything from your backup system while the primary site is being repaired.
Some might say that if downtime mattered in the olden days, we just used replication – and that provided something similar to instant recovery. While that is true, it represented an additional system and service that many companies did not have the money for. As a result, most companies stuck to tapes and a vaulting service for DR – and hoped they never had to use them.
These and other reasons are why companies are now seeing DRaaS as a strong alternative to whatever they were doing before. The latest example of this is Bluelock – a company known previously as a hosting company for VMware’s vCloud – that has pivoted strongly toward DRaaS. In fact, it has done this so strongly that Gartner lists Bluelock as one of very few leaders in that space.
Bluelock is a multifaceted company. It is part consulting company, part reseller, and part independent software vendor. It’s DRaaS offering starts with allowing you to use most DRaaS-enabling tools that makes sense for your organization. This includes using tools to assist with non-Intel offerings like iSeries. (Most DRaaS vendors will support only Windows & Linux on Intel.) The person briefing us said it’s harder to find a requested tool Bluelock doesn’t support. If you’re unsure which tools to use, its consulting arm will assist you in finding the best tools for your needs. Bluelock is setup as an MSP with these companies, allowing it to give you pay-as-you-go pricing to use those tools in your DRaaS setup.
If you view Bluelock as just a reseller, though, you will miss the biggest advantage of using its service. Its developers created a management and reporting portal that runs on most any device, which allows you to see exactly how your DR setup is performing. It even allows you to push a single button on your mobile device to declare a disaster. Supporting multiple DR tools is one thing; having an application that manages and reports on them is another. It’s quite impressive indeed.
Bluelock is quite an interesting company. It is one of the only vendors listed as a leader by Gartner that does not sell its own DR product. Bluelock allows you to pick which tool best works for you, including tools that support non-Intel environments. Since it’s really hard to find a single software product that everyone likes, this “reseller way” of looking at things allows Bluelock to appeal to a much broader customer base than the typical DRaaS offering. It also better matches the real data center, which always has an application, running on an odd server that is still critical to operations and need inclusion in any DR plan.