Today, Amazon Web Services is a mainstay in the cloud infrastructure services market, a $ 60 billion corporate giant. But in 2008 it was still new and was working to stay afloat and cope with the growing demand for its cloud servers. In fact, the company launched Amazon EC2 in beta 15 years ago last week. From that point on, AWS started offering unlimited computing power, a major selling point at the time.
EC2 was one of the first real attempts to sell elastic computing on a large scale – that is, server resources that scale as needed and disappear when they don’t. As Jeff Bezos said in an early sales presentation to startups in 2008, “You want to be prepared for lightning strikes. […] because if you are not, it is really going to cause a great deal of regret. When lightning strikes and you weren’t prepared for it, it’s hard to live with. At the same time, you don’t want to prep your physical infrastructure for some kind of hubris in case lightning doesn’t strike. So, [AWS] somehow helps in this difficult situation. “
A first test of this value proposition came when one of their startup clients, Animoto, scaled from 25,000 to 250,000 users in four days in 2008, shortly after the company’s Facebook app launched on South by Southwest.
Back then, Animoto was a consumer-targeted app that allowed users to upload photos and turn them into a video with a background music title. While this product may sound tame today, it was then state-of-the-art and used a significant amount of computer resources to create each video. It was an early portrayal not only of user-generated Web 2.0 content, but also of the connection between mobile computing and the cloud, which we take for granted today.
For Animoto, which launched in 2006, choosing AWS was a risky proposition, but the company found that trying to run its own infrastructure was even more of a gamble due to the dynamic demand for its service. Starting up your own server would have required enormous investments. Animoto first went that route before turning to AWS because it built before it received initial funding, explained Brad Jefferson, co-founder and CEO of the company.
“We started building our own servers because we thought we had to prove the concept with something. And when we started doing that and got more traction from a proof-of-concept perspective and started letting certain people use the product, we took a step back and thought, well, it’s easy to prepare for failure, but what we need to prepare for success, ”Jefferson told me.
Choosing AWS may be an easy one given what we know today, but in 2007 the company really put its fate in the hands of a largely unproven concept.
“It’s pretty interesting to see how far AWS and EC2 have come, but it was really a gamble back then. I mean, we spoke to an e-commerce company [about running our infrastructure]. And they try to convince us that they will have these servers and it will be fully dynamic and so it was nice [risky]. In hindsight it seems obvious, but it was a risk for a company like us to bet on them back then, ”Jefferson told me.
Not only did Animoto have to trust AWS to do what it said it could do, but it also had to spend six months redesigning its software to run on Amazon’s cloud. But when Jefferson analyzed the numbers, the choice made sense. At the time, Animoto’s business model was free for a 30-second video, $ 5 for a longer clip, or $ 30 for a year. When he tried to model the amount of resources his company would need to get the model up and running, it got really difficult. So he and his co-founders decided to bet on AWS and hope it would work if and when there was a surge in usage.
That test took place at South by Southwest the following year, when the company launched a Facebook app, which spiked demand and, in turn, pushed the boundaries of what AWS was capable of at the time. A few weeks after the startup launched its new app, interest exploded and Amazon began looking for the right resources to keep Animoto going.
Dave Brown, who is now Amazon’s VP of EC2 and was an engineer on the team in 2008, said that “everyone” [Animoto] Video would initiate, use, and terminate a separate EC2 instance. In the previous month they had used between 50 and 100 instances [per day]. Their usage peaked at around 400 on Tuesday, 900 on Wednesday and 3,400 on Friday morning. ”Animoto was able to keep pace with the increasing demand and AWS was able to provide the necessary resources. Usage eventually peaked at 5000 instances before settling down, which in the process proved that elastic computing could actually work.
At that point, however, Jefferson said his company didn’t just trust EC2’s marketing. There have been regular phone calls to AWS executives to ensure their service does not collapse under this increasing demand. “And most importantly, can you get us more servers, we need more servers. To their credit, I don’t know how they did it – when they took the processing power off their own website or others – but they were able to get us where we needed to be. And then we were able to get over that peak and then things naturally settled down, ”he said.
The story of keeping Animoto online became a major selling point for the company, and Amazon was actually the first company to invest in the startup alongside friends and family. It raised a total of $ 30 million, with the last funding coming in 2011. Today the company is more of a B2B operation helping marketing departments make videos easily.
While Jefferson didn’t provide details on the cost, he did point out that the price of trying to maintain servers that would be dormant most of the time was not a viable approach for his company. Cloud computing turned out to be the perfect model, and Jefferson says his company remains an AWS customer to this day.
While the goal of cloud computing has always been to deliver as much computing as you need, whenever you need it, these particular circumstances put that notion to the test.
Today, the idea of having problems generating 3,400 instances seems strange, especially when you consider that Amazon is now processing 60 million instances every day.