Amazon reported that the outages were triggered by one so-called “networking event” in its East Coast facility that caused a cascade of problems that led to crashes, taking down major sites like Reddit, Foursquare and Hootsuite. Amazon has not revealed how many companies use its cloud computing services or how many were affected by the outage.
Not surprisingly, the event is prompting all industry observers to comment on the reliability of cloud computing and has sparked much debate about the future of cloud computing and what needs to happen for companies to continue to put their trust in cloud vendors such as Amazon. In a report in the New York Times, IDC analyst Matthew Eastwood called the outage a “wake-up call for cloud computing” and said “It will force a conversation in the industry.”
That discussion, he said, will most likely center on what data and computer operations to send off to the cloud and what to keep inside the corporate walls. But another issue, Mr. Eastwood said, will be a re-examination of the contracts that cover cloud services — how much to pay for backup and recovery services, including paying extra for data centers in different locations. That is because the companies that were apparently hit hardest by the Amazon interruption were start-ups that, analysts said, are focused on moving fast in pursuit of growth, and less apt to pay for extensive backup and recovery services.
The fact is that, while unfortunate for Amazon to be sure, such an event was inevitable given the widespread adoption of cloud-based strategies and the wide range of companies and popular sites that rely on it. IT professionals considering a migration to the cloud certainly should pay close attention to the details of and fallout from an event like this, and the directive to focus on backup and recovery is smart advice. To call an outage like this the end of cloud computing would be catastrophic in itself: It is, as Eastwood put it, simply a wake-up call to the reality of technology that will prompt the development of solutions and safeguards and, ultimately, make the cloud model even more resilient.