Netflix is now 100% all-in on the AWS cloud

Video-streaming service Netflix has completed its seven-year push to reduce its reliance on private datacentres so it can run all its operations in the Amazon Web Services (AWS) cloud.

Yury Izrailevsky, vice-president of cloud and platform engineering at Netflix, confirmed in a blog post that the company had completed its wholesale migration into the AWS cloud in January 2016.  

A report in the Wall Street Journal revealed in August 2015 how close Netflix was to going all-in on the AWS cloud, with the article suggesting the move would be completed by the end of the summer.

In the blog post, Izrailevsky said the cloud move took as long as it did because shifting Netflix’s infrastructure off-premise took a lot of hard work and required the company to make some “difficult choices”.

The firm had to completely rewrite a lot of its processes and applications so they would run natively in the cloud. It also had to change the way its teams operate, so they could adopt a continuous delivery strategy for new software code.  

“Many new systems had to be built, and new skills learned,” he said. “It took time and effort to transform Netflix into a cloud-native company, but it put us in a much better position to continue to grow and become a global TV network.”

While this has been going on behind the scenes, Netflix has signed up 75 million subscribers and rolled out its services to 130 countries, with the help of AWS’s portfolio of global datacentres.

“Supporting such rapid growth would have been extremely difficult out of our own datacentres; we simply could not have racked the servers fast enough,” said Izrailevsky.  

“Elasticity of the cloud allows us to add thousands of virtual servers and petabytes of storage within minutes, making such an expansion possible.”

He went on to describe the service availability benefits Netflix has received since starting the process of downsizing its datacentre footprint in August 2008.

Outages were a common occurrence when operating its own datacentres, and do still occur in the cloud, he admitted, but the move off-premise has resulted in greater reliability overall.

“Failures are unavoidable in any large-scale distributed system, including a cloud-based one,” he said. “However, the cloud allows you to build highly reliable services out of fundamentally unreliable but redundant components.

“By incorporating the principles of redundancy and graceful degradation in our architecture, and being disciplined about regular production drills using Simian Army, it is possible to survive failures in the cloud infrastructure and within our own systems without impacting the member experience.”

The move to cloud has also enabled the company to cut costs, said Izrailevsky, which he described as more of a side benefit to the move.

“This is possible due to the elasticity of the cloud, enabling us to continuously optimise instance type mix and to grow and shrink our footprint near-instantaneously without the need to maintain large capacity buffers,” he said. “We can also benefit from the economies of scale that are only possible in a large cloud ecosystem.”