On June 13, Amazon Web Services (AWS) experienced a minor outage that lasted for about three hours. At 12:08 pm PDT, the company reported that it was “investigating increased error rates and latencies” in some parts of the United States. The outage affected many mainstream news organizations, such as the Associated Press, which were unable to publish articles.
No Impact on Ethereum Network Nodes
Ethereum advocate Evan Van Ness noted that despite the fact that the Ethereum network largely relies on Amazon’s hosting, it was not affected by the outage. According to Ethernodes, 64.5% of the Ethereum network depends on Amazon hosting providers. Van Ness added that the impact might have been more significant if the outage had occurred in Europe due to the amount of Ether (ETH) staked on Lido, currently around 7.1 million or 35% of the total. He said, “I imagine there would be some effect if AWS went down in Europe given how much of Lido is in the cloud.”
Centralization Criticism
Ethereum has previously been criticized for centralization due to its reliance on infrastructure provider Infura, which provides network nodes to companies and organizations. Many of these companies, as well as the liquid staking platform Lido, rely heavily on AWS for cloud hosting services.
Root Cause of the Issue
Around 20 minutes after the problem was discovered, AWS stated that the root cause of the issue was related to a service named AWS Lambda, which allows customers to run code for various kinds of applications. More than three hours after AWS went down, the company announced that “the issue has been resolved, and all AWS Services are operating normally” at 3:37 PM PDT. According to hosting platform Kinsta, AWS has the largest market share among cloud hosting providers, at 34%.