Electrical Storm, Leap Second Bug Shake Up the Web: Online Businesses Beware


Over the weekend, online businesses like Instagram, Pinterest, and Reddit suffered unexpected problems, the result of a severe electrical storm that took out a critical data center, and a “leap second” bug, the product of attempts to correct the world’s atomic clock. As businesses increasingly use collective resources within the cloud, your company could be in more danger than ever of becoming a casualty.

Amazon Outage

The perfect storm. Over the weekend a string of online titans, including Amazon’s cloud customers Netflix, Instagram, and Pinterest, experienced outages when a single Amazon data center in Northern Virginia was disrupted by an electrical storm and backup systems failed to function. All Things Digital

A knock out punch. Here’s the blow by blow as Amazon’s Elastic Compute Cloud went crashing down Friday, taking a string of high profile sites with it. Netflix and Pinterest remained down until well after 9 p.m. PT, while other services did not seem to fully recover until early the following morning. Venture Beat

Working Out the Bugs

Hold on a second. As if Friday’s fireworks were not trouble enough, the Web experienced another technical glitch Saturday evening as software handling operations for sites like Reddit, Mozilla, and Gawker hiccuped on a so-called “leap second” added recently to the world’s atomic clock to put it in sync with the Earth’s rotation. (Who knew?) Wired

Stop bugging us! If an electrical storm caused Amazon Web Services’ outage problems Friday evening, it’s looking like some previously unknown bugs in the system prolonged it. If the strength of cloud services is shared resources, this might also be its weakness. It’s important for businesses to know the risks of hosting in the cloud. ZDNet

Forecast Cloudy

Down with the ship. A bit of speculation has some experts asking how to use the cloud’s benefits without suffering its misfortunes. Many businesses depend on cloud services for diverse operations, but this can also make them vulnerable. TechCrunch

Prescient predictions. Almost a month before this weekend’s problems, Yale University researcher and assistant computer science professor Bryan Ford was warning that growing interdependencies within the cloud could lead to instabilities and even potential meltdowns. His research and the recent problems might make any business owner think twice. PC World

Silver Linings

On second thought. The outage at Amazon Web Services has many businesses taking stock. While time and effort are certainly saved by leasing software and computer power from a cloud provider, companies find they have little control over situations when technical problems arise. The Wall Street Journal

Getting up to speed. If you’re just now looking into cloud computing for your business, all this talk probably has you a bit worried. Cloud computing holds some uncertainty, true, so it might help to learn more about what the cloud is and what it can mean for your company. Josh Lowry

For Better or Worse

Don’t get blown away. Once you’ve decided to move some of your computer operations to the cloud, there are some risks you must understand. From service continuity issues to e-mail security concerns and much more, cloud computing means certain sacrifices in return for the benefits obtained. CloudTech

A winning strategy. Perhaps the most obvious solution for dealing with outages in cloud provider services is the one many large scale companies are already pursuing. It’s time to follow a multi-vendor solution for your cloud computing needs. SocalTech.com

1 Comment ▼
One Reaction
  1. Anita Campbell

    And it wasn’t just huge sites and large corporations that experienced “leap second” problems — our BizSugar.com site was almost brought down due to that problem last Friday night. We didn’t know what the issue was until SoftLayer made us aware.