A few months ago, Amazon scored a major hire. It convinced Adrian Cockcroft to come work for the company as VP Cloud Architecture Strategy.
Cockcroft had been consulting for Battery Ventures, helping the VC find cloud startups to invest in.
But he’s best known for his years at Netflix as Amazon Web Services’ most famous first customer. He led the project, way back in 2009, to have Netflix build out its movie streaming service on AWS and not use its own data centres. That was an insane decision at the time. Cloud computing back then was mostly known as a crummy, unreliable and perhaps unsafe alternative to owning your own computers. When, in 2010, Netflix started to publicly talk about that decision, the world thought Netflix was being somewhere between idiotic and reckless.
“I did a talk at a conference at the end of 2010 to 100 people or so. The reaction of the audience: ‘You guys are crazy,'” Cockcroft told Business Insider last year. “[AWS] was small. It was unreliable. All kinds of things weren’t there. We were basically helping create what’s there now by making it clear to AWS, ‘if you do this, this will work,’ and ‘this doesn’t work,’ and ‘we need this feature’.”
Even by 2012, AWS wasn’t exactly reliable. It famously went down on Christmas Eve that year, making national news.
Flash forward to early 2017 and the world loves AWS and all cloud computing and companies are racing to put their businesses on it. Cockcroft’s main job at AWS is to talk to companies doing what Netflix did back in the day — get rid of their data centres and go all in on AWS.
A growing list of companies are doing that such as Intuit, Time, Juniper, AOL, Hertz and more all all the time, he explained on stage at a conference in San Francisco on Tuesday.
But sometimes the downside of relying on cloud services like Amazon’s can become painfully evident.
And unfortunately that was the case on Tuesday when at the very moment Cockcroft was on stage making the case for AWS, a good chunk of the internet had been taken down (including Business Insider’s site) because AWS’ computer storage service, S3, was suffering major technical problems.
The service went down so thoroughly that AWS was even having trouble updating the “health dashboard” that tells customers if its service is up or down. A message on that page read (emphasis ours):
“We have now repaired the ability to update the service health dashboard. The service updates are below. We continue to experience high error rates with S3 in US-EAST-1, which is impacting various AWS services. We are working hard at repairing S3, believe we understand root cause, and are working on implementing what we believe will remediate the issue.”
Even after it updated that page to indicate there was a problem, AWS very distinctly didn’t call it an outage. It called it “high error data rates.”
In the few minutes while Cockcroft spoke onstage touting the advantages of Amazon’s cloud services, likely unaware of the AWS technical troubles occurring that very moment, an internet meme was born.