Most organizations understand the need for effective disaster recovery (DR). However, the cost and complexity of traditional DR solutions often force small to midsize businesses (SMBs) to take unnecessary risks.

“SMBs generally have been unable to take advantage of traditional DR solutions for a variety of reasons, the primary one being cost,” said Vince Conroy, CTO, FusionStorm. “Another problem is that these companies don’t have the depth of expertise around technologies such as data replication and failover automation that are required to implement a disaster recovery solution.

“In addition, these organizations typically don’t have a secondary facility in place — someplace that’s at least 70 miles away from their primary site. They can always rent co-location space but that’s not very cost-effective, and they may not have the staff to maintain and test the DR site. Because application environments are constantly changing, a DR site is only as good as the last time you tested it.”

To address these challenges, FusionStorm has introduced a family of managed DR solutions that provide organizations with comprehensive DR capabilities backed by FusionStorm’s engineering expertise. These solutions are customized based upon the customer’s DR needs and objectives, and take advantage of the latest virtualization, automation and replication technologies to maximize reliability while minimizing costs.

At Your Service

FusionStorm’s managed DR solutions work in conjunction with its IT Xpress managed hosting service to provide customers with a DR site in one of FusionStorm’s data centers. Depending upon the customer’s needs, FusionStorm will implement the DR site using physical or virtual servers, a dedicated storage array or shared utility storage, data replication services and ongoing management. The replication element can be simple or complex depending upon the number and type of applications to be replicated and other factors.

“We support a wide variety of data replication technologies, including EMC Recover Point, which is an enterprise-class replication technology that supports heterogeneous environments — it doesn’t matter what systems or storage you have on either end. We also support proprietary, array-based replication from EMC, Hitachi and other vendors,” Conroy said. “For smaller environments we support several host-based and appliance-based replication technologies, including DR Scout from InMage.”

Customers don’t have to worry about implementing these technologies or keeping them up and running. FusionStorm handles everything, and provides 24x7 monitoring and problem resolution for the communication circuit linking the primary and DR sites. Most importantly, FusionStorm performs ongoing testing of the recovery process.

“We not only spin up the environment and make sure it works but we do regular testing — typically biannual — where we coordinate with the customer to test the failover from production to recovery site and verify that all of the applications are running,” said Conroy. “If a customer already has a secondary data center we can manage that infrastructure for them, although it’s usually more cost-effective and efficient for us to do this in our Sacramento or Las Vegas hosting facilities.”

Holding Down Costs

FusionStorm is able to make these solutions cost-effective through economies of scale and technologies that help reduce the cost and improve the reliability of the DR environment. Recovery servers are deployed as virtual machines insofar as possible to minimize the amount of hardware required. In addition, the virtual machines are placed in “quiescent mode” when they’re not in use.

“The DR server only needs to be running when we’re doing a test or there’s an actual disaster,” Conroy said. “We can reduce costs if we build that server as a virtual machine and then quiesce it – it remains in an inactive state stored on a disk until needed. This reduces both hardware and power requirements.”

Some customers choose to utilize a portion of their DR site resources for testing, development, staging or sandbox functions. With virtual machines it’s simple to create such a dual-purpose site.

Depending upon the number of servers and applications involved, FusionStorm engineers will also use various automation solutions to help handle the failover from production to the recovery site. These automation products include VMware Site Recovery Manager, EMC AutoStart and InMage DR Scout.

Customized Solution

FusionStorm’s managed DR solution begins with a consulting engagement called Discovery. FusionStorm engineers help each customer complete a questionnaire that defines the customer’s specific DR requirements, including recovery point objective (RPO) and recovery time objective (RTO).

“Those are the two most critical pieces of information,” Conroy said. “RTO is the maximum amount of time a customer can tolerate from the moment a disaster is declared until the recovery site is available. It can be as short as an hour for some customers and as long as five business days for others, although typically we’re doing sub-four-hour RTOs. RPO reflects the maximum amount of data loss a customer can tolerate — that is, the volume of data that has not yet been replicated to the DR site at the moment a disaster occurs.

“Based upon the RTO and RPO, along with the applications involved and the type of storage the customer is using, we can determine what DR technologies we’re going to use and what the cost is going to be to the customer.”

FusionStorm understands the challenges SMBs face when it comes to disaster recovery, and has leveraged its extensive engineering expertise and data center investments to create a comprehensive DR solution. Although many SMBs have nonexistent or inadequate disaster recovery solutions due to the high cost and complexity of traditional DR technologies, FusionStorm’s managed DR solution reduces business risk by providing end-to-end protection that’s cost-effective and simple.