The roster of clients is modest, but Dart’s Disaster Recovery and Data Centre is immense, sophisticated and considers itself among the region’s leading data-protection facilities.
Purpose-built in 2007, the multimillion-dollar Forum Lane unit is category 5 hurricane-rated, providing redundant power, security, data communications connectivity and seven private disaster recovery suites with 40 workstations customers can use to maintain business continuity.
The suites, according to Dart IT department operators, “house employees and staff where workstations are ready for use, [and] phones and IT equipment are activated for network connectivity.”
His/her bathrooms, shower facilities and a kitchenette and conference room complete the facilities.
“There are a couple of other facilities on island that offer similar capabilities,” says Nadege Parent, Dart Realty marketing manager for client services, “but the Data Centre’s strengths are the robustness of the physical building, a back-up cooling system and 24-hour security. It’s one of the most technologically advanced and physically secure of its kind in the Caribbean.”
Its communications infrastructure, she says, is designed for international banking, financial services, legal, insurance, accounting and retail businesses.
As robust as any disaster-recovery network is, however, a business owner ultimately hopes never to have to use it at all, preferring comprehensive continuity and data-protection services.
In fact, leading computer security research and education organization, the SANS Institute points out that “the most successful disaster recovery strategy is one that will never be implemented.”
SANS, officially known as the Escal Institute of Advanced Technologies, says never using a disaster recovery plan means “risk avoidance is a critical element in the disaster recovery process.”
The group offers some startling statistics: “Only 15 percent of midrange data centers would be able to recover more than 30 percent of their applications in any time frame; just 3.8 percent could recover their applications within the same day; only 2.5 percent could recover within four hours.”
As a counterpoint, in late July, the Basel, Switzerland-based Committee on Payments and Market Infrastructure, a 25-member group of central banks, said exchanges, banks, brokers and other institutions must be able to restore clearing houses, payment systems, trade repositories, and clearing and settlement houses within two hours of any cyberattack.
The goal is ambitious, and the SANS statics indicate full recovery may be unachievable, and that makes data protection and resiliency all the more critical.
Information availability is key to the success of most businesses, says Parent. The Data Centre’s back-up, recovery, co-location and managed-hosting services allow continuity of computer and network operations, providing “routed and switched infrastructure.”
The description of the Data Centre’s IT services is likely to baffle most business owners, but a rough translation is the network is able to use multiple routes to move information to a destination that is secure – and probably distant – from outages, storms or other emergency events.
“The network layout promotes client and network segmentation,” says Parent, meaning software can segment email, telephone and website data, then send it through a variety of “virtual local area networks.”
“That allows clients to set up their own independent, managed networks – for example,” she says, “between an office in Camana Bay and overseas.”
The description continues: “The network consists of an external public facing routing table and an internal facing routing table, offering 180 routed switch ports for client access.”
The phrases simply mean that the network has a list of destinations for each client’s information and a route map to get it there.
That network, Parent says, is “a fiber-to-the-premise solution,” providing high speed and high bandwith, “a superior communications infrastructure,” suited to the kinds of high-profile clients that use the Data Centre.
Japan’s Sumitomo manufactures the air-blown fiber, called “FutureFlex,” which, according to its website, allows scalability – network capacity and shifting bandwidth requirements – and efficient use of conduit space, lower costs and the ability to upgrade as new technologies become available.
Data Centre clients will keep excellent company with such FutureFlex users as the Pentagon in Washington, D.C., Cable News Network and sports programmer ESPN, Johns Hopkins University, the Mayo Clinic, at least two international airports, Intel and Nissan.
The 4,421-square-foot Data Centre, Parent says, uses the MAYA-1 and Cayman-Jamaica Fiber System submarine cables; it boasts dedicated, onsite technicians, 24/7 monitoring of customer equipment and infrastructure, 10 co-location racks and 25 customer-dedicated racks, redundant communications connections with Cayman’s four service providers, redundant air-conditioning units, fire suppression and environment controls, and redundant power supplies and backup generators.
“The availability of this commercial infrastructure,” Parent says, “lowers the cost of doing business in the Cayman Islands as companies wishing to provide internal systems applications and online client services do not need to bear the cost of creating [their own] data center.
“The disaster recovery suites offer the ability to co-locate all critical business functions in a facility which can be used full time or part time” at competitive costs, which she pegs provisionally – “we tailor the fees to individual requirements” – at US$1,500 for a half rack of servers, and $3,000 for a full rack.
“They can be leased on a short-term basis, e.g. before, during or after a storm.
“The facility provides a much-needed service for companies requiring a cost-effective business infrastructure for nonstop computing,” she says.
Data Centre facilities fall nicely in line with SANS recommendations, which start with a “crisis management plan,” dependent on “the ability to handle high-level coordination activities surrounding any crisis situation.”
Development, maintenance and testing of a disaster recovery plan is next, and that involves precisely the kind of external routing the Data Centre provides: “To be strongly considered,” SANS says, “is a recovery strategy for alternate processing,” which it calls a “hot-site.” Any recovery plan must identify a hot-site “if the primary location is not available to provide disaster recovery services for the various system environments.”
Parent is careful to stipulate that Data Centre does not itself perform data recovery, that the center is a “secure physical space to locate a server providing redundancy in the event of a natural disaster,” but “back-up and data recovery are the responsibility of the client.”
The center offers the requisite technology and services; using them is up to the client.
Mindful of security, Parent hesitates to name Data Centre clients, although she enumerates “around 10,” and offers a testimonial from law firm Ogier, headquartered at Camana Bay’s 89 Nexus Way.
“We went through Hurricane Ivan in the Cayman Islands in 2004, so we really wanted to move somewhere where we had secure backup power, particularly for our IT connections,” said partner James Bergstrom.
In the event corporate disaster recovery becomes necessary, says the SANS Institute, “recovery activities will be conducted in a phased approach,” starting with activation of a disaster recovery plan, moving operations to a back-up site and an emergency operations center within 24 hours.
Phase 2 seeks to reverse the process, “to recover critical business functions, restor[e] critical applications and critical network connectivity.
“The goal here,” SANS says, “is to recover the systems and network so that … customers can continue business.”
Phase 3, finally, is “return[ing] data-processing activities to the primary facilities or another computer facility.”
In sum, a successful recovery effort comprises “restoration of critical applications to the most current date available on backup tapes stored off-site.
“It is understood that, due to the emergency or disaster, response times will probably be slower than normal production situations,” SANS says. “The plan provides recovery procedures to be used at the present data center site after repairs have been made or at the disaster recovery back-up site and the emergency operations center.”
The institute offers a final caution to businesses, saying, “Those that have adequately developed, maintained and exercised their contingency plans will survive,” and warns corporate executives who “take the uninhibited operations of their companies for granted.
“They remain complacent, assuming that the power will always be available, the telephone system will not fail, there will be no fire or earthquake – everything will always be normal.
“The final corporate contingency plan is the lifeblood of corporate survival. A fundamental premise of successful contingency planning is that plans are developed by those who must.”