Dating site dumps Amazon EC2: “100% uptime a required SLA”
The storm which accounted for Amazon’s cloud outage over the weekend has resulted in one company ditching the AWS system over fears of further unavailability.
From 2300 EST on June 30 users of several high profile sites such as Pinterest, Netflix and Instagram were unable to access content, with AWS “investigating connectivity issues”. The issue was resolved by 0054 EST on June 31 but updates on power restoration appeared sporadically throughout the day.
Amazon apologised for what had happened in a statement which read: "We know how critical our services are to our customers’ businesses. If you’ve followed the history of AWS, the customer focus we have, and the pace with which we iterate, we think you know that we will do everything we can to learn from this event and use it to drive improvement across our services".
It appears, however, that for one company this downtime was the final straw.
WhatsYourPrice.com, an online dating service which allows users to arrange dates by naming their price, has gone its own way with Amazon following the provider’s latest EC2 outage. Reports state that WhatsYourPrice.com has now upped sticks and moved to Las Vegas-based cloud provider FiberHub.
Perhaps the most worrying aspect of this was that, according to WhatsYourPrice, Amazon's support team didn't respond to their calls to a previous outage on June 14.
WhatsYourPrice.com CEO Brandon Wade said that the outage had hit the reputation of his company, stating that 100% uptime was “a required service level agreement for anyone providing cloud computing services”.
“Services focused on dating and relationships require constant accessibility,” he said, adding: “Dating is all about the serendipity of meeting the right person at the right time. If an online dating service is not available, a user may lose the chance to meet his or her soul mate forever”.
Despite the fact he put it in a starry-eyed romantic way the WhatsYourPrice owner does have a point – so can the cloud be trusted?
According to calculations from the International Working Group on Cloud Computing Resiliency (IWGCR), cloud services overall were only up for 99.917% of the time – far less than the holy grail of 99.999% availability, or even the 100% uptime Wade spoke of.
Yet this research may not be all it appears – the stringency of the IWGCR’s methodology was called into question after they took the majority of their material from press cuttings. “The procedure to gather information is far from exhaustive”, they admitted.
With the IWGCR hoping to utilise Google Alarms among others for future reports, watch this space for a more exhaustive look at cloud resiliency next year.
It could be argued that the cloud can be trusted as much as any computing service – in other words, it will have very occasional periods of downtime but that’s the nature of the beast.
But what do you think? Did Wade have a right to be angry at Amazon for this latest outage when the original fault appeared to be out of their hands?
- » What can MSPs gain from linear growth and recurring revenue?
- » Cloud disaster recovery in a nutshell: How does it work?
- » Rackspace posts solid financial results, continues to bang drum for managed cloud services
- » Here’s why VMware hasn’t left it too late with its hybrid cloud push
- » Google launches PerfKit, open source benchmarking tool for cloud performance