If you run IT for a smaller midsized company, what kinds of affordable disaster recovery services are out there for you? We asked several leading DR providers about their managed disaster recovery and business continuity services for companies with up to about 200 employees. All of them offer disaster recovery services tailored to this segment.
With prices starting at $100 a month, we found several options for organizations with a monthly budget in the hundreds of dollars, as well as a few priced in the low $1,000s. The difference? How fast you get access to your data after an outage, what you personally have to do to make that happen, and what else comes as part of the service, such as temporary access to applications hosted by the provider. Here's an overview of options from two traditional DR providers, IBM and SunGard Availability Services LP. Watch for additional coverage from other types of providers in the coming days.
The most basic tier, starting at $100
The least expensive option we found for managed disaster recovery services was SunGard's basic recovery services. Customers of this service back up their data however they like -- to tape, disk, even iPods. Then, if they have a system failure, they take that data to the provider's site and restore it themselves on the provider's hardware. Travel is required, sometimes long distances. The recovery time objective (RTO), or time to restoration, is 48 hours or longer.
For companies that can stomach that RTO, buying standard recovery services at a secondary site is "a perfectly viable" option, said Jim Olson, vice president, solutions engineering and training at SunGard Availability Services. SunGard's monthly fees for Intel servers can range from $100 to several thousand dollars, depending on the number and type of Intel servers, Olson said. Larger platforms like Sun Solaris or HP Superdome servers cost more. Tip: Make sure you have a vendor that has the platform you need to recover to.
What you get for $500 a month
This tier is also offered by SunGard. One option is infrastructure management, which is essentially out-tasking production functions to SunGard rather than handling them internally. An infrastructure management solution, which includes space and power and no minimum rack commitments, starts at $500 per month. (Full cabinet/dedicated solutions start at $1,500 per month for standard hosting.)
Another option is disk-based recovery, usually over a network, with an RTO of 24 hours. The RTO is shorter because the data in a disk-based environment typically is more readily available. Monthly fees start at $500 and increase based on the size of the data and scope of the recovery. "If you do a full advanced recovery, which would be renting colo space at a secondary site, with dedicated assets at that site and a network connection to restore several terabytes of data, the cost would be $10,000 to 20,000 a month," Olson said. In this situation, RTOs are usually well under 24 hours, he said.
When you have $1,000 a month
IBM has both managed services and infrastructure recovery services (which include a business continuity component) starting at $1,000 a month.
The infrastructure recovery services provide a secondary site for recovery of systems and people, said Brian Reagan, IBM's director of business continuity and resilience services. They span the classic hot-site options: a place to mirror the application or system to a secondary site. Customers must make sure IBM has hardware that can accommodate their data requirement: all the shared storage, systems management assistance and testing to bring those systems and applications back into operation.
Fees range from $1,000 to $2000 per month, based on the applications and size of the workforce. IBM has some midmarket companies running mainframes, "which have more moving parts and are more expensive," Reagan said.
The workforce recovery component is based on the number of people requiring online access to the recovery center. "In a 200-person company, not all 200 need to be in a secondary location at the time of disaster," Reagan said. "At 50 seats, you're looking at probably about $1,000 a month for having a secondary location."
Buying dedicated seats -- meaning access any time of day or night -- costs more. You can also buy shared seats. "We will essentially guarantee that you will have access to your seat on a scheduled basis at time of disaster, which means we have to make sure we are not over-syndicating and we can accommodate all the requests that might come into a specific center," Reagan said. Tip: Grill providers on how much syndication they are doing to ensure against being left out in the cold in a regional disaster.
Managed services -- the infrastructure and support provided by IBM during a disaster -- run about $1,000 a month for companies with 200 or fewer employees and a terabyte of data, Reagan said. As prices fall, the capacity seems to be settling at $1 to $2 per gigabyte of service.
For $2,000 a month
SunGard offers server or disk replication, where data is replicated in real time between devices, starting at this price point. With RTOs of less than 12 hours and in most cases less than six hours, replication's other benefit is a recovery point objective, or data loss, of zero. Monthly fees start at $2,000 a month.
If you have more than $2,000 a month
IBM offers consulting services for disaster recovery planning in this price range. For a small business, a consulting engagement might be a two-week project where IBM defines the critical applications and devises a business continuity, workforce and crisis communications plan, Reagan said. These engagements can run from several thousand dollars to $25,000, depending on the duration of the project.
Coming next: A reseller's DR services; Savvis looks forward to DR in the cloud.
Let us know what you think about the story; email Linda Tucci, Senior News Writer.