Hurricane Irene Heads for Amazon Web Services Data Center

Via Cbowns on Flickr

Amazon acknowledged that its Virginia-based data center was in the path of Hurricane Irene, which could cause potential outages for several high-profile websites.

Geekwire discovered an Amazon representative’s reply in the AWS forums addressing concerned clients who had data stored in the Northern Virginia data center.

The representative said the company was monitoring the hurricane and making all possible preparations, including generator fuel, food/water, flashlights, radios and extra staff.

The representative also recommended checking out the AWS Architecture Center, which offers several tips on how to build highly scalable and reliable applications in the AWS cloud.

This same Virginia-based data center failed in April 2011, knocking many big websites like Reddit, Quora, and Foursquare offline.

When that happened, we interviewed Adam Ely, director of security at TiVo, and he offered some valuable suggestions to help minimize the impact in disaster events.

Ely recommended that companies prepare for disaster by hosting their websites in multiple data centers across the United States.

With Irene bearing down on Amazon’s Virginia center, the reasoning behind his recommendation becomes crystal clear. Spreading your hosting needs over Amazon’s centers in California, Virginia, and Ireland would minimize impact in this scenario. Going beyond that, you could even opt to differentiate your providers: like hosting 10 servers with RackSpace and 10 with Amazon.

Ely did point out one drawback with the multiple-hosting locations: cost. You will pay more to host in various data centers.

Comments have been disabled for this post.
Sort: Newest | Oldest
badler 5 pts

Natural disasters such as Hurricane Irene expose an inconvenient truth – an AWS region can be viewed as a single point of failure. You may have redundant power, networking, and isolated infrastructure, but if an entire geographic area is affected, your cloud region could very well be a casualty of the event. As such, it is desirable (a necessity?) to either spread your application architecture across multiple regions/clouds, or have a disaster recovery environment in another region or with another infrastructure provider. While a true multi-region architecture looks good on paper, in practice it is complicated by bandwidth costs, latency between regions, and security issues (once your bits and bytes leave the region, they are in the Wild West of the public Internet). It is for this reason that a lot of customers we talk to come into a conversation convinced they need a multi-region/cloud environment, but usually leave with a plan to make the best use of availability zones and have a solid DR solution in place.

- Brian Adler, RightScale PS Architect