When organizations have investigated Disaster Recovery in the past, the necessity to basically double the infrastructure was a show stopper for 90% of decision makers. While the solution of full DR was enticing, the financial implications were too drastic, which shot down the concept of DR in executives’ minds for a number of years. Microsoft has realized this, and with their ever evolving strategy and cloud capabilities, Azure Site Recovery is a feasible DR solution. This offering has put the idea back on the table for small to midsize organizations and has made DR a reality.
Dennis Junk of Aptera has written a terrific article on how this new offering is a game changer not only within DR solutions, but for Microsoft in general in their direction and willingness to play with others. Below is the article Dennis wrote:
DR in Theory and DR in Reality
What makes this a big deal is that backing up infrastructure is a costly and complicated endeavor for all kinds of businesses. Disaster recovery may seem like it should be a relatively simple matter, especially when you’re using virtual machines. In principle, it should look like this:
For whatever reason, the infrastructure hosting your business’s workloads ceases to function properly. Fortunately, like all prudent business leaders, you’ve provisioned back-up servers to take over the hosting in case of just this eventuality. If everything goes according to plan, as soon as the primary system crashes, your back-up system starts up, and your workloads are switched over. You may experience a brief amount of downtime before your software is up and running again. But, with any luck, your mission critical processes will continue with only a minor disruption.
But of course the reality is usually much more complicated. And Microsoft sees this as an opportunity to step in and address a real challenge that countless businesses are facing.
Eric Rupp, the Infrastructure and Cloud Services Practice Leader here at Aptera, gives the example a big company with offices in both Fort Wayne and Indianapolis. The Fort Wayne office relies on an environment comprised of about 500 physical servers. In the event of a system failure in Fort Wayne, a system switches critical functions over to back-up infrastructure housed in the Indianapolis office. This means that the backup machines have to be provisioned, maintained, and ready to go at all times. Keeping this many servers at the ready—even though most of the time they’re not actually being used—is a time-consuming and costly challenge.
In addition to the costs associated with the back-up hardware, the company also has to pay licensing fees for the failover software that actually switches from one set of machines to the other. For this particular company, Rupp estimates the licensing fees are in the range of a million dollars annually. And he says this isn’t an especially high figure when compared to other companies’ situations.
Azure Site Recovery services that failover to customer-owned infrastructure run $16/ month/ instance. (An instance is another name for a virtual server.) So, if you have 500 servers, you’re looking at $96,000 annually. That’s quite a savings from a million. If you’re just setting up your disaster recovery system and you don’t have a back-up server farm, you have Azure Site Recovery failover to Azure’s own cloud infrastructure. Failover to Azure is $54/ month/ per instance.
The really appealing thing about cloud back-ups is that you can potentially save quite a bit on server provisioning and maintenance. It costs more once you’re switched over to Azure and running off of Microsoft’s datacenters, but you only ever pay for what you use. Once you’re system is repaired, you go back to your own datacenters and go back to paying the $54 a month per server. And you don’t need to purchase, configure, and maintain a second server farm.
Azure also allows you to build an orchestration script. After a system-wide failure, you’ll want to prioritize getting the applications that are most critical to your business back online. An orchestrated recovery follows your script for what order to concentrate on when returning functionality to all your business’s systems.
Playing Well with Others
VMware has a disaster recovery service that fails over to the cloud as well. But so far neither Amazon Web Services nor Google Cloud Platform offers anything like Azure Site Recovery. It’s less of a surprise these days than it was a few years ago that Microsoft is building technologies that work well with tools from other companies. The writing’s been on the wall for quite a while now that Windows isn’t the only game in town, so they can’t exactly expect people choose products that are going to give them fewer options.
What makes the general availability of Azure Site Recovery interesting is that disaster recovery seems like such an obvious area for cloud providers to focus on—and yet Microsoft is getting out ahead this time. By opening the service up to VMware clients, the company is broadening the base of potential new customers by a huge margin. This could make it difficult for the Amazons and the Googles to catch up. And Microsoft may also be thinking that once they have their foot in the door with disaster recovery tools, they may be able to push through with other products in their cloud lineup.
To see Dennis’ article in its entirety, please visit https://apterainc.com/.
If you would like to hear how Azure Site Recovery can help your organization, please reach out to firstname.lastname@example.org and someone will reach out to you shortly.
Martin Feehan, PEI
If you’re interested in learning more about Azure or how Azure can help to streamline your business and save you money while giving you increased computing power, attend PEI’s 7th Annual Sporting Clay Event on Azure Workloads and Cloud Computing!