On May 7, a sudden spike in temperature at one of Amazon's data centers in Northern Virginia caused a power outage, disrupting cloud services. Today, the company reported that most operations have been restored. In particular, the affected crypto exchange Coinbase has already resumed trading.
On May 7, a sudden spike in temperature at one of Amazon's data centers in Northern Virginia caused a power outage, disrupting cloud services. Today, the company reported that most operations have been restored. In particular, the affected crypto exchange Coinbase has already resumed trading.
Thursday's outage was the second major incident in recent months caused by overheating. Earlier, in November last year, the CME Group derivatives exchange suffered one of the longest outages in recent years due to a cooling system failure in CyrusOne's data centers, Reuters writes .
As of 8:12 a.m. Eastern Time (ET), the number of AWS outage reports on Downdetector had dropped to 72, down from a peak of nearly 600 on Thursday night.
AWS said it is ramping up additional cooling capacity, but the process of adding the resources needed to safely restore all remaining systems is taking longer than expected.
The cloud platform also said it has redirected traffic for most services from the affected “Availability Zone.” Such a zone consists of one or more interconnected physical data centers and is designed to operate autonomously within an AWS region.
Last October, AWS experienced a massive outage that caused global disruption to thousands of websites, including popular apps like Snapchat and Reddit.
It was the largest cyber incident since the CrowdStrike outage in 2024, which paralyzed hospitals, banks, and airports, and highlighted the vulnerability of interconnected technologies in today's world.
