Amazon's cloud service division, AWS, said a service outage that lasted approximately 15 hours on Monday (October 20) has been resolved. The incident highlighted the global internet's heavy reliance on a single company's cloud services.

The world's largest cloud computing provider said AWS services had "returned to normal operation" as of 6:00 p.m. New York time on Monday (October 20).

AWS holds approximately one-third of the global cloud computing market. According to website monitoring platform Downdetector, the outage affected the internet services of hundreds of companies, including financial services platforms Venmo and Robinhood Markets, Apple's music and TV services, Zoom Communications, software companies such as Salesforce and Snowflake, restaurant giant McDonald's, and gaming company Epic Games. Amazon's voice assistant Alexa and home security system Ring were also affected.

Corey Quinn, chief cloud economist at cloud spending consulting firm Duckbill Group, said this was likely AWS's worst outage since a major outage in December 2021. "The question is, is this a major incident? Or is it a result of our increased interconnectedness and reliance on Amazon?" he said.

AWS earlier said a glitch in the digital directory of a critical database service prevented software that relied on it from retrieving information, triggering a cascading outage. The company said it had located and fixed the underlying issue by Monday morning New York time. The outage primarily affected AWS's operating region on the US East Coast, its largest data center cluster.

However, during the repair process, engineers discovered that other subsystems were also affected by the database outage, including a critical component used by customers to launch newly leased servers.

Most outages in large technology systems are typically quickly repaired. However, the high interconnectedness of technology systems means that a problem at one company can have a global impact. Last year, a botched software update by cybersecurity firm CrowdStrike grounded flights and caused system outages worldwide, resulting in billions of dollars in losses.