Rawpixel - Fotolia

AWS datacentre service trouble hits Netflix users

North Virginia datacentre hit by service trouble, blighting delivery of more than 30 of the cloud giant's products

Amazon Web Services (AWS) suffered technical difficulties over the weekend which blighted the activities of a range of cloud services that are run out of its North Virginia datacentre.

Technical problems were reported on more than 30 of the firm’s cloud services on 20 September 2015, including its app monitoring tool CloudWatch, its NoSQL database offering DynamoDB, and its real-time data processing tool Kinesis.

The service problems affected the output of its North American datacentre for several hours, with the DynamoDB downtime blamed for follow-on periods of disruption at Netflix, Reddit, and Amazon’s own Instant Video service.

Computer Weekly contacted AWS for comment on the problems, but was advised to consult the firm’s online status tool for further information about the cause and nature of them.

At the time of writing, the dashboard suggests all problems have now been resolved, with all of the affected services reported to be operating as normal.

Speaking to Computer Weekly, Clive Longbottom, service director of market watcher Quocirca, said that despite the attention downtime episodes like this garner, would-be cloud buyers should not let it dissuade them from moving their IT off-premise. 

"Let's imagine Netflix built its own datacentres, and the same data issue happened. How long would it take to get things up and running again? Would Netflix be able to afford the skilled staff AWS has in place? Possibly, for a company of that size, but small firms wouldn't," he said.

"That's the comparison that counts: How much outage would there be in a private datacentre? How quickly would such a facility take to get back up and running? What skills would be needed to do so? 

"Reports about downtime at AWS might put some off, particularly those who have always been in the 'I told you so' camp, but it shouldn't put off anyone else. Cloud is still a major part of any IT platform's future," Longbottom added.

Past problems at AWS

The company’s North Virginia datacentre is responsible for serving up cloud services to North American users, and has been the source of past outages at the firm, including two concurrent periods of service trouble in August 2013.

The North Virginia datacentre was the first one the company opened, going live in 2006, and since then AWS has opened further facilities across the globe – including Ireland, Frankfurt and South America – to provide users with access to local cloud services.

AWS is not the only tech giant to experience technical difficulties recently, as users of the Microsoft-owned messaging service Skype have been beset with service issues for much of today (21 September 2015).

As reported by Computer Weekly, this has left users of the Skype app unable to send and receive calls or see if their contacts are online or not. The browser-based version of the service is not thought to be affected.

Read more about AWS

Read more on Infrastructure-as-a-Service (IaaS)