Recent catastrophic events have cast a new light on cloud services and co-location providers. The question that must be asked is this, what happens if the cloud provider or colo goes down?
In New York City, building and fire codes mandate that fuel be stored in the basements of buildings, so many data centers also house fuel pumps, tanks, and generators in the same location. When Hurricane Sandy hit in 2012, some data centers went down when those basements flooded, including some colocation providers.
In the wake of a disaster, the failsafe has to work, and Hurricane Sandy exposed a possible risk for those using cloud or colocation providers. According to Mark Evanko, principal engineer at BRUNS-PAK, co-location or cloud sites are less expensive than owning and operating multiple data centers, but if a catastrophe occurs at a co-location center, the outside provider does not take responsibility for the cost of service interruptions. Moreover, co-location sites can expose tenants to security risks if data are exposed or stolen.
Redundant facilities owned by one organization are expensive, but this option provides data center managers with more control and security. Another option is to use multiple co-locations, either by employing one provider with multiple sites or by using more than one provider.
Cloud technology uses virtualization, whereby data along with their applications can be easily moved from one server to another anywhere in the world. “Virtualization is helpful,” says Mark Levy, analyst, data centers, for 451 Research, “but mission critical facilities need 99.9 percent uptime, and it is not always appropriate to virtualize every application, because it may not perform as well.”
Go here for more information on how data center and facility managers can use cloud services, co-location, and redundant facilities to safeguard the data center in the wake of a catastrophe.