Amazon datacentre glitch hits top web services

A hardware failure at Amazon datacentre downs several web services, renewing concerns about the reliability of cloud-based infrastructure

A hardware failure at an Amazon datacentre has downed several top web services, raising renewed concerns about the reliability of cloud-based infrastructure.

Instagram, Netflix, Twitter's Vine video-sharing application and holiday site Airbnb were among the services that were slow or inaccessible at the weekend because they all rely on Amazon Web Services.

The fault was traced to a datacentre in northern Virginia that was struggling to keep up with demand, according to the BBC.

Other services hit by the fault included cloud-based software firm Heroku and web automation service IFTTT.

After the problems were resolved, Amazon said the interruptions to service was caused by the "partial failure of a networking device" but gave no further details.

Read more about Amazon Web Services (AWS)

  • Cloud scalability on Amazon Web Services calls for creativity
  • AWS alternatives: What else is there but Amazon Web Services?
  • Amazon Web Services cuts dedicated instance pricing for market grab
  • Startups omit in-house IT with Amazon Web Services
  • Three things you didn't know about big data and Amazon Web Services (AWS)
  • Lamborghini slashes web costs with Amazon Web Services

The fault was linked to a network-attached block level storage service called Elastic Block Store (EBS) that is used for applications that need a database, file system or access to raw block level storage, according to the Register.

The weekend interruptions came less than a week after Amazon's North American web stores went offline for about 30 minutes, also due to problems in the company’s northern Virginia datacentre.

It followed a similar disruption on Google that lasted four minutes and reportedly caused a 40% drop in global internet traffic.

The service interruptions have highlighted the vulnerability of cloud-based services, showing that glitches at single datacentre can cause widespread disruptions.

Read more on Business continuity planning