When the Cloud Fails: Steven Meyer on the Global Ripple Effect of the Amazon Web Services Outage (AWS Outage)
Earlier this month, RTS – Radio Télévision Suisse reported on a major outage that disrupted much of the global web. A failure within Amazon Web Services (AWS Outage) that left major platforms such as Snapchat, Fortnite, Airbnb, and Netflix offline for several hours.
In the broadcast, ZENDATA Co-Founder Steven Meyer shared insight into how centralised cloud infrastructure can amplify the impact of such incidents. He also discussed whether local data centres could help reduce dependency on foreign providers.
Meyer’s comments underscored a growing concern across industries: the world’s digital backbone is concentrated in the hands of a few global players. His perspective reflects ZENDATA’s ongoing commitment to cyber resilience, regional autonomy, and responsible cloud practices. The principles that safeguard both business continuity and national digital sovereignty.
Below is the full English translation of the RTS report, published on 6 November 2025.
How an Amazon Outage Paralyzed Part of the Global Web
Several popular websites, online games, and applications — including Snapchat and Fortnite — were inaccessible for several hours on Monday after an outage in the United States at Amazon Web Services (AWS), Amazon’s cloud platform. The incident illustrated the dependence of global businesses on American tech giants. A new wave of disruptions appeared later that evening before the problem was finally resolved.
Dozens of websites and applications such as Airbnb, Reddit, and video games like Fortnite, Roblox, and Brawl Stars were intermittently unavailable. In Switzerland, services from UBS, Swisscom, and Sunrise were also affected.
Applications at the heart of millions of daily routines faltered after a failure around 9 a.m. Swiss time at an AWS data centre in Virginia, near Washington D.C., demonstrating the world’s reliance on the infrastructure of large U.S. technology companies.
See Hackmanac’s cybersecurity post on X showing the scale of affected sites.
New Disruptions in the Evening
The situation seemed to improve by lunchtime in Europe. Around 12:30 p.m. Swiss time, AWS indicated on its maintenance page that “most AWS service operations are now functioning normally.”
However, subsequent maintenance updates caused additional complications, impacting new clients even as others, such as Lloyds Bank in the U.K. and the AI start-up Perplexity, announced that their services were back online.
According to Downdetector, a site that tracks connection issues, a resurgence of outages hit more users around 3 p.m. GMT, affecting Battlefield, Delta Airlines, and Netflix in the U.S.
Problem Finally Resolved
Shortly before midnight in Switzerland — roughly fifteen hours after the crisis began — AWS announced the issue had been completely resolved, with full recovery expected “within two hours” as pending workloads were processed.
A subsidiary of Amazon, AWS is the world’s largest cloud computing platform, providing companies with shared data centres, private servers, and artificial intelligence tools. It holds nearly a third of the global market — ahead of Microsoft Azure and Google Cloud, which together account for another third, according to Synergy Research Group.
Initially, AWS attributed the outage to a DNS issue, the system that routes internet traffic to its correct destination. But later, a more serious cause emerged. Around 6 p.m. Swiss time, AWS confirmed that “the root cause was an internal subsystem responsible for monitoring the health of network load balancers.” In other words, the issue wasn’t just navigation — it was a control tower failure.
Local Servers?
Interviewed Monday on the RTS Forum programme, Steven Meyer, co-founder of ZENDATA and cybersecurity specialist, explained that although the internet was originally designed as a fully decentralised network, cost pressures have driven companies to centralise data into large data centres and cloud systems.
“Amazon operates many data centres and cloud services that are supposed to mitigate this kind of issue,” Meyer said. “But replicating data across multiple regions is expensive and complex. So, most people store their data in a single region.”
When asked about the possibility of storing data locally to prevent such widespread consequences, Meyer said that while local data centres are an option, they don’t eliminate the risk entirely.
“Outages happen everywhere — even Swiss data centres have experienced them,” he explained. “Choosing a local option doesn’t change that risk directly; it just avoids dependency on a third-party service based in the United States.”
