On a recent Monday, a significant service disruption at Amazon Web Services (AWS), the powerful cloud provider underpinning a vast portion of the internet, brought numerous websites and applications to a grinding halt for more than two hours. This event served as a fresh reminder of the inherent vulnerability within our interconnected global digital infrastructure.
Beginning shortly after 3 a.m. Eastern, the widespread outage impacted critical services for major banks, popular gaming platforms, and various entertainment apps. By 5:27 a.m., Amazon announced that most services had been restored to normal operation, though they were still diligently addressing a backlog of pending requests.
The list of affected popular services was extensive, including communication giants like WhatsApp, essential government websites and tax platforms in Britain, the payment app Venmo, and the cryptocurrency exchange Coinbase. Even digital games offered by The New York Times and The Wall Street Journal’s website faced interruptions. Numerous other major companies and retailers, from Amazon itself and Hulu to Snapchat, McDonald’s, Ring doorbells, and the popular game Fortnite, reported significant service disruptions.
By 9:45 a.m. Eastern, internet outage tracker Down Detector had logged over eight million reports concerning Amazon Web Services, with a substantial number originating from users in the United States and Britain.
At 11:43 a.m., Amazon disclosed that their preliminary investigation pointed to an internal system, responsible for managing network traffic distribution, as the root cause of the widespread service issues in Northern Virginia.
Industry experts quickly highlighted that this incident once again exposed the internet’s precarious dependence on a handful of dominant tech providers, such as Amazon, Microsoft, and Google. When one of these critical services falters, millions of users worldwide experience immediate disruptions. They recalled a similar, even larger, day-long internet blackout just last year, triggered by a flawed software update from the cybersecurity firm CrowdStrike.
Thousands of clients trust Amazon Web Services for their most intricate, demanding, and data-heavy operations, encompassing everything from streaming high-definition video and powering complex web applications to securely storing colossal volumes of digital data.
Harry Halpin, CEO of NymVPN, a virtual private network service, suggested that Monday’s problem likely originated from a technical glitch within one of Amazon’s primary data centers. However, he emphasized the inherent lack of transparency in cloud platform operations, making it impossible to ascertain the exact cause without an official disclosure from Amazon.
Dr. Halpin recounted waking up to a flurry of emails from soldiers on the Ukrainian front lines, whose VPN services his company provides, inquiring about the cause of the disruption. He noted that this critical reliance on cloud services extends to numerous governments worldwide.
“When an entire nation’s digital infrastructure is concentrated with a few U.S.-based providers, and these services can fail at any given moment, whether due to malicious attacks or simple technical errors, it creates an incredibly perilous scenario,” Dr. Halpin warned.
“People have grown accustomed to the internet’s reliance on a select few providers,” Dr. Halpin observed, “but this dependence is far from normal and carries significant risks.”
Amro Al-Said Ahmad, a computer science lecturer at Keele University in England, explained that Amazon’s ‘us-east-1’ region in Northern Virginia, the reported epicenter of Monday’s issues, houses one of its most massive data centers. While acknowledging that ‘cloud computing generally functions well for daily operations,’ he cautioned that even a minor misstep, such as a faulty software update, possesses the power to bring down an entire system.
The outage, which notably interrupted secure communication apps like Signal and other vital digital tools, prompted calls from media advocates for an urgent move towards greater diversification in cloud computing infrastructure.
In a powerful statement, Corinne Cath-Speth, head of digital for Article 19, a prominent free speech advocacy group, asserted that ‘the foundational infrastructure supporting democratic discourse, independent journalism, and secure communications simply cannot remain dependent on a mere handful of corporations.’
Curiously, despite the widespread disruption, Amazon’s stock price showed little movement in premarket trading, indicating that investors were largely unfazed by the incident. This perhaps reflects the massive importance of AWS; in the first half of the year, Amazon Web Services generated almost 20 percent of Amazon’s total sales, yet contributed approximately 60 percent of its operating profit.
Amazon’s formidable cloud-computing division boasts an expansive global infrastructure. Clients leverage AWS services to seamlessly scale their operations up or down as needed, sidestepping the massive capital investment and maintenance associated with building and managing their own hardware.
Rebecca Wright, a computer science professor at Barnard College in New York City, pointed out that smaller businesses particularly benefit from entrusting their digital needs to cloud providers like Amazon Web Services, which possess specialized expertise. She acknowledged the trade-offs but argued, ‘I wouldn’t advise companies to avoid outsourcing cybersecurity to experts in the field.’
Mehdi Daoudi, founder of internet performance monitoring firm Catchpoint, reflected on how two decades ago, companies typically managed their own data centers. Today, he noted, the majority depend on cloud services from giants like Amazon, Google, Microsoft, or Chinese providers. Interestingly, however, the escalating costs of these cloud solutions have recently led some businesses, over the past two to three years, to reconsider and revert to managing their own internal infrastructure.
This recent outage could intensify existing calls for both businesses and governments to prioritize and utilize cloud services hosted within their respective regions.
Alexandra Geese, a German Member of the European Parliament, responded to Monday’s disruption by advocating for crucial European data and digital infrastructure to be hosted exclusively within Europe, by European companies, and under strict E.U. jurisdiction.
She emphasized that the outage served as ‘a stark reminder that Europe’s digital sovereignty is not an abstract concept, but a vital matter of security and resilience.’
Additional reporting was contributed by Melissa Eddy in Berlin, Jeanna Smialek in Brussels, and Andrés R. Martínez in Seoul.