Users have voiced their concerns that Amazon's EC2 cloud computing services has been slowing down.
One affected user, Alan Williamson, who co-founded software company aw2.0 which specialises in deploying software within cloud networks, said that normally the network response from Amazon EC2 (known as a ping) takes about 0.3 milliseconds. "We have discovered that when our [application] instances appear to be dying or at least shaky, then this network latency jumps up to a whopping 7,241 milliseconds [or seven seconds] to move a network packet around internally."
Cloudkick, a company which specialises in cloud performance monitoring, has investigated Amazon's slowdown. It found that the network times were more than just blips, with major latency drag lasting for more than a few hours at a time.
The company ran a sample of ping latency tests across several hundred EC2 application instances. The results are shown in the graph below.
Commenting on the results, Cloudkick said, "The spikes in latencies up to 1, 000 milliseconds are definitely abnormal, and should never be encountered on healthy private network,"
Amazon said, “Overall, we are not having over capacity issues as has been reported in some areas. We sometimes work with customers to tweak their configurations or it could mean making modifications in our services to assure maximum performance.”