Amazon's cloud computing services have experienced performance problems this afternoon, with multiple services affected. There are also numerous reports of users briefly being unable to access the main Amazon.com retail site.
Amazon's Service Health Dashboard showed problems on the EC2 computing cloud in the US. "We detected a period of elevated packet loss from 12:31 PM PDT to 12:46 PM PDT in a single Availability Zone," Amazon reported. "We are continuing to monitor the situation." The dashboard also showed elevated error rates on its Amazon's CloudFront CDN, SimpleDB database service and Mechanical Turk freelance marketplace. The downtime was also confirmed by monitoring services CloudStatus and enStratus, which both show the Amazon services available again as of 2 pm Pacific.
The outage is the second in a month for Amazon Web Services, following a June 11 incident in which a lightning strike damaged power equipment at one of the company's data centers, disrupting service for some AWS customers. Today's problems come at a time of growing scrutiny of the reliability of cloud computing providers. EC2 previously experienced extended outages in February 2008 and October 2007.