BACKGROUND IMAGE: iSTOCK/GETTY IMAGES
Devising and testing disaster recovery requires a great deal of time and effort. DRaaS, or cloud DR services, are gaining traction among enterprises because they offer lower cost and greater flexibility. While DRaaS addresses shortcomings with existing on-premises DR processes, it has a few warts of its own.
Traditional disaster recovery requires a corporation to identify which applications it wants to bring back up and then put redundant data center infrastructure -- servers, software, network connections, storage -- in place. The business also must standardize procedures so employees understand how to invoke backup processes and make up-to-date backup information available to workers. All of this comes at a price.
If a major natural or manmade disaster occurs, there may not be enough room at the inn to run every company's DR applications.
Spending hundreds of staff hours and, in some cases, more than seven figures to mitigate a 1% -- or even a 10% -- risk doesn't seem like a good return on an IT investment. Traditionally, vendors built and maintained autonomous backup data centers that usually sat idle.
"Cost has been a disaster recovery issue," said Karyn Price, industry analyst, cloud computing services at San Antonio, Texas-based consulting firm Frost & Sullivan.
Cloud-based DR services address some of the long-standing market challenges. Disaster Recovery as a Service (DRaaS) lowers the cost threshold; cloud services run on shared infrastructure, so the vendors' investment is less, and they pass the savings along to customers. Advancements in virtualization and automation enable suppliers to offer customers more flexibility; companies can sign up for services for all of their applications or for only select services and use them as needed.
Testing in the cloud is another plus. Because it was such a complicated process, businesses tended to test their DR plans sporadically -- once or maybe twice a year. With cloud, businesses are able to test their services regularly. The needed infrastructure sits ready to go, so many hurdles are cleared, putting the impetus on a company to determine the amount of time they want to invest in testing.
Vendor shortcuts create DRaaS downfalls
DRaaS vendors include a variety of testing options with their services. In some cases, they offer unlimited testing, but in other cases, it is similar to traditional services -- the more tests conducted, the higher the fee.
DRaaS is available from horizontal cloud suppliers, such as Amazon Web Services, iland Inc., IBM, RapidScale and Verizon, as well as DR specialists such as Axcient Inc., Bluelock, EVault, Net3Technology, PHD Virtual Technologies, QuorumLabs Inc., SunGard, Windstream Communications and Zerto.
Because of the benefits, interest in DRaaS has been rising. Gartner Inc. expects worldwide cloud DR revenue to increase from $640.8 million in 2013 to $5.8 billion in 2018 -- a 55.2% compound annual growth rate.
But there may be some limitations with the services.
"Vendors do not have complete system redundancy," said Rachel Dines, a Forrester DR analyst.
It does not make sense for suppliers to build data centers that mimic all of their customers' infrastructure, so they take a shortcut: They build their systems so that they can handle a limited number of outages. Theoretically, a business should be able to recover its system if it encounters a site-specific problem, such as an electrical outage in its data center. However, if a major natural or manmade disaster occurs, there may not be enough room at the inn to run every company's DR applications. What's worse, a company would only find out if they're supported when the disaster has already occurred. Essentially, DRaaS may not be available if a catastrophic disaster occurs.
About the author:
Paul Korzeniowski is a freelance writer who specializes in cloud computing issues. He has been covering technology issues for more than two decades; is based in Sudbury, Mass.; and can be reached at firstname.lastname@example.org.