At every industry conference, manufacturers and service providers have been preaching the benefits of the growing cloud computing phenomenon. It will be the preferred way for corporate America to operate, they have crowed.
And then, there was Amazon. The Internet retailing giant also provides cloud computing services to a large number of clients. And in April, Amazon Web Services went down for four long days.
The cause? A configuration error was made during a network upgrade. In other words, a seemingly minor human error led to outages for several well-known web sites including Quora, Foursquare and Reddit. Unlike many web server issues, this was not a slowdown in service or occasional 404 message. Instead, entire companies were taken down over an entire weekend.
Almost immediately, the naysayers of the cloud computing concept began coming out of the woodwork. They began arguing that the idea of off-site hosting, storage and applications just doesn’t make sense for the enterprise.
Instead of this knee-jerk reaction, it would do IT professionals good to analyze the situation and see what can be learned – and how cloud computing can be safer, better and more reliable in the future.
For its part, Amazon published a highly transparent post-mortem of the entire situation.
Some of its customers are being quite public and forthcoming with their own experiences as well. For example, Netflix published a report on its experiences over the outage weekend. Netflix’s customers ran into more errors than usual, but the company’s web site and services generally remained live. This has only reinforced, however, Netflix’s decision to remain in the cloud.
“We set out to build a highly available Netflix service on AWS and this outage was a major test of that decision,” Netflix said in a published statement. “While we do have some lessons learned and some improvements to make, we continue to be confident that this is the right strategy for us.”
Like Netflix, I believe that companies who are considering or already rely on cloud-based computing services could ultimately become stronger based on the lessons learned from Amazon’s failure. First and foremost is this: every company needs a disaster recovery plan that extends to its unified communications system, data storage, servers and more.
Most firms that keep their servers for data and apps on-premise also rely on some kind of off-site backup. And those who don’t should start this practice today. Similarly, when a company chooses to operate in the cloud, IT personnel must have a complete understanding of potential failures and weaknesses in this strategies. Redundancy, backup and fail-safes are the key.
This isn’t the end of cloud computing. It’s just a growing pain. And over time, the same advantages that have made cloud computing a technology buzzword of 2011 will prompt its development into a stable, reliable platform for corporate America.
About Tom Kincannon
