Webmetrics reported today that Amazon S3 maintained more than 99.99 percent uptime for the month of October, exceeding the performance requirement of its recently introduced SLA. Webmetrics monitored the response times and availability for the REST- and SOAP-based APIs for Amazon S3 during October, using its GlobalWatch Monitoring platform. Performance results for each API were as follows:We're hearing of more and more startups using Amazon S3 (Simple Storage Service) for their storage needs; and this stat will please both them and their users.
|Web Service||Average Uptime||Average Response Time|
|Amazon S3 REST API||99.9915%||1.63 sec||Amazon S3 SOAP API||99.9912%||1.55 sec|
S3 has more than five billion objects currently under management; including for 37Signals, webmail.us, YouOS, ElephantDrive, adaptiveblue, and more. Indeed just this week PollDaddy, the poll app we use at R/WW, reported they'd switched to S3. Polldaddy told us that "the system [S3] is very stable and all polls will now be served off Amazon S3 so there should never be any lag."
Starting October 1st, Amazon introduced a new SLA for S3 guaranteeing users 99.9 percent service uptime. If uptime is less than 99 percent, customers can apply for a service credit of 25 percent of their total S3 charges for the month. If the uptime is 99 percent but less than 99.9 percent, customers can apply for a service credit of 10 percent of their monthly charges.
Once again we're seeing the benefits of Amazon's HaaS (Hardware as a Service) strategy. As Emre Sokullu said in his article, Amazon is leveraging its deep scalability know-how and expertise. It's making web publishing even easier and cheaper - and reliable. This is a real game changer on the Web, especially in this new age of utility computing.