What the Huge AWS Outage Reveals About the Internet

A massive cloud outage stemming from Amazon Web Services’ (AWS) key US-EAST-1 region, its hub in northern Virginia, near the US Capitol, caused widespread disruptions of websites and platforms around the world on October 20. Amazon's main ecommerce platform and other properties, including Ring doorbells and the Alexa smart assistant, suffered interruptions and outages throughout the morning, as did Meta's communication platform WhatsApp, OpenAI's ChatGPT, PayPal's Venmo payment platform, multiple web services from Epic Games, multiple British government sites, and many others. The outages stemmed from Amazon's DynamoDB database application programming interfaces in US-EAST-1, and AWS said in status updates that the problem was specifically related to domain name system (DNS) resolution issues. AWS has suffered other large-scale outages, including a major incident in 2023. Reliance on central cloud services from giants like AWS, Microsoft Azure, and Google Cloud Services has, in many ways, improved cybersecurity and stability around the world by creating a baseline of guardrails and best practices for all customers. But this standardization comes with major trade-offs, because the platforms become a single point of failure for large swaths of critical services.