The Washington PostDemocracy Dies in Darkness

Amazon Web Services outage hobbles businesses

Amazon's widely used cloud service, Amazon Web Services, experienced a large-scale outage on Nov. 25. (Video: Reuters)
Placeholder while article actions load

SEATTLE — Amazon's widely used cloud computing service suffered a major outage in its eastern U.S. operations Wednesday, hampering everything from services for Web-connected security cameras to software applications that businesses use to design products.

Starting midmorning, a variety of Amazon Web Services applications began to fail, including ones that deliver data and authorize users to access that data. That hobbled companies that rely on AWS, including the Amazon-owned Ring security camera service, iRobot’s Roomba vacuum cleaner app, services from design technology firm Autodesk and the publishing systems of news outlets such as The Washington Post (which is owned by Amazon founder and chief executive Jeff Bezos).

AWS is the world’s largest provider of cloud-computing services, which let customers rent data storage and processing capabilities over the Web instead of running their own data centers. AWS, which competes in the business of providing computing infrastructure via the cloud against Microsoft and Google, has emerged as Amazon’s most profitable business.

Amazon held 45 percent of the global market in 2019, according to the market research firm Gartner. Companies as diverse as Netflix, Kellogg’s and Airbnb run significant pieces of their computing operations on AWS. Amazon’s dominance also comes with a price: When its operations go down, they can take out wide swaths of the Web, as Wednesday’s outage showed.

Amazon’s next closest cloud-infrastructure competition, Microsoft, held 17.9 percent of the market last year, Gartner said.

As Amazon wrestled to resolve the outage, it offered little in the way of explanation.

“We are working toward resolution,” Amazon spokeswoman Mary Camarata said.

A few hours after the outage began, Amazon wrote on the AWS status page that it had “identified the root cause” of the problem and was working to resolve it. By midafternoon on the East Coast, Amazon had updated its status report, saying it “expect(s) full recovery to still take up to a few hours.”

The outage appeared to be limited to the collection of data centers Amazon runs that are scattered over a handful of Northern Virginia cities, according to the AWS status webpage. That data-center grouping, known as a “region” in cloud-computing lingo, is the oldest in Amazon’s operations, having launched in 2006.

The impact of the outage was felt broadly. New account activation and the mobile app for streaming media service Roku were hampered, even as the streaming service continued to work on televisions, spokeswoman Tricia Mifsud said via email. And the Target-owned Shipt delivery service could receive and process some orders, though it was “taking steps to manage capacity” because of the outage, spokeswoman Danielle Schumann said in an email. Photo storage service Flickr tweeted that customers couldn’t log in or create an account because of the AWS outage.

The AWS disruption took down some features, such as the ability to log in or watch video, of the Ring security camera system, Ring spokeswoman Yassi Shahmiri confirmed. Amazon owns Ring. In a tweet, iRobot said that the AWS outage was “currently impacting our iRobot Home App” that manages its Roomba vacuums. Autodesk noted on its product status webpage that the AWS outage was “impacting multiple” services and applications from the design software company.

News sites such as the Wall Street Journal and the Chicago Tribune also noted that the outage affected their operations.

Amazon is best known for its massive e-commerce site. But AWS is the business that generates the most profit for the firm. In the last quarter, AWS posted $11.6 billion in sales, 12 percent of the company’s overall revenue, but $3.5 billion in operating income, or 57 percent of its profits.

Amazon pioneered cloud computing, launching AWS in 2006 to offer companies the ability to run their technical operations on the company’s vast array of servers. The business helped spark tech start-ups who could rent computing as they needed it, rather than laying out large sums of money to purchase their own hardware and software. Later, bigger corporations turned to cloud computing as it proved its value.

A massive AWS outage two years ago disabled large chunks of the Internet, including sites such as Slack, Medium and Quora. That outage, which also hit the Northern Virginia region, was caused by human error, when an employee mistyped a command, taking down huge portions of the system and knocking out AWS’s core storage service.

The latest outage struck just a week before AWS’s annual conference, which will be held online this year, rather than in person in Las Vegas, where it typically can be found.