After Sandy: Companies Still Fail to Adopt Cloud

Hurricane Sandy affected companies and organizations that rely upon the Internet to continue their businesses, but questions have been raised regarding additional means of protection against similar disasters.

For websites and services which take full advantage of the cloud's flexibility and redundancies, cloud-based deployment appears to be the most obvious solution to this problem.

Michael Bloomberg, the mayor of New York, acknowledged the problem of having so many Internet service providers located within one geographic area, including the fact that a lot of Internet resources are put at serious risk. As flooding knocked out power, fuel supply pumps and back-up generators, many data center operators and their customers were affected negatively by this gamble.

Equinix and Internap, larger Internet providers, were able to run on either utility power or stable generator power, and things, by Saturday, were beginning to get back to normal. This recovery, however, was not uniform in other areas or with other organizations; Datagram, a company which hosts websites like Huffington Post, Gawker and BuzzFeed, had no access to its basement generators and only acquired a back-up generator on Friday afternoon.

When Sandy hit a week ago, 2,000 separate sites on the Internet using unique IP addresses were cut off. There were great efforts made to restore and keep data centers' connectivity to the Internet following the disaster. However, even if these efforts do pay off, the same problem will still exist; and another storm may threaten to wipe everything once again.

Sandy did not make companies realize the significance of utilizing off-site public clouds, even as simply a back-up option for online presence. No company, it would seem, is safe from the possibility of a disaster, and they should therefore be able to keep going when a problem arises, or be able to get back up to work once it has been knocked down. Websites work on networks, and consequentially do no have to be to one vulnerable server in one vulnerable building.

However, it is not simply enough to tell companies to put their sites on the cloud; websites need to be designed to be specifically movable. Complying to regulations will also be a factor for some industries, as will the cost of replicating data over many locations.

Robert Offley, the CEO of managed service provider CentriLogic, believes that the cost is not a big issue. The replication price for a site five years ago, according to Offley, was “six to eight times higher than now.” Site replication is therefore a more cost-effective option in today's day and age.

There are many other cloud-based solutions, including site replication, that can help companies make the most of the cloud. Other options would be backing-up the site's database in an off-site facility or developing a cloud architecture that works across many locations in a multi-site capacity; this would allow websites to run smoothly from server-to-server depending upon regional network and data center outages.

When compare to the loss of website traffic or revenue which can affect a business when a problem comes up, the price of adopting one of the many cloud-based solutions is not very expensive at all.

Following the events after Sandy, many businesses and tech-savvy organizations have failed to put the time and effort into making active back-ups of their data. It is a particular shame that cloud technology is not being adopted by organizations at the moment, especially because it prevents a website from failing due to an unforeseen problem in a single location.

Critics may well argue that clouds are not perfect solutions; they do not offer duplication and high-availability services out of the box, and can be affected like any other data center. Web admins and developers, amongst other customers, need to develop sites which can take full advantage of the cloud; the danger otherwise is that the server will be affected negatively if the building it is sitting in gets knocked offline for any reason.

Overall, web teams are predicting success in the cloud, and flexible services offer a more up-to-date method in avoiding and dealing with disasters. (CY) Link