At this week’s ASW re:Invent 2013 conference in Las Vegas, John Furrier and Dave Vellante welcomed Brennan Saeta, Infrastructure Engineer with Coursera. Coursera is an education company that partners with the top universities and organizations in the world to offer courses online for anyone to take, for free.

Furrier noted the disruptive quality of Coursera, as it could basically be cutting in on the business model employed by higher-education institutions. However, Saeta quickly pointed out the collaborative relationship Coursera shares with their university partners worldwide.

“There is still a lot of value in what the universities provide and obviously we wouldn’t exist without our university partners. What we want to do is help our university partners expand from simply teaching the students who they can attract and who can afford tuition and receive financial aid and really expand beyond so they can teach people who don’t really have the resources, the means or the time to be able to come online.”

There were obvious challenges for Coursera in its early days. But those challenges have ultimately brought the company to what it is today, a free education site that serves over five million students, moving more than a petabyte of data monthly in the process.

Coursera came to AWS in the fall of 2011 when, under the guidance of Stanford faculty wanting to run an experiment, they began working with Amazon. “And we haven’t looked back,” says Saeta. “It’s been phenomenal. Amazon has allowed us to scale our capacity in ways that we never thought would be possible.”

A company in the cloud

With the exception of the laptops used by the company’s developers, every other aspect of Coursera is located in the cloud.

The development team, until only a few months ago, consisted of only two full time engineers. This despite the massive scale they are currently managing. “Our students are more active than you would see in other environments,” Saeta commented. “Even though the number of students we have isn’t as much as some of the other big websites out there, there’s a lot of load and a lot of students and we simply can’t manage all of the servers individually. So we very quickly moved to having everything be in autoscaling groups.”

This, according to Saeta, allows them to shoot a machine in the head and watch the data come back immediately.

Saeta has a blog he has detailed some of the work done for Coursera at betacs.pro

With a commitment to Amazon, Coursera’s primary data store is Amazon RDS. But they use a full spectrum of AWS products for their mission. “We don’t care how it gets done, it just needs to be done quickly,” he said.

Vellante was curious how Coursera has opted to back up and protect all of that data. Saeta informed him they store in S3, which has massive availability. They also utilize Glacier, though not as much as they should, Saeta admitted. Coursera feels comfortable with their data in RDS due to their daily incremental backups.

Coursera is excited by the RDS release of the cross-region copy/snapshots. “We want to back stuff up outside of the US East region that we currently reside in,” commented Saeta.

Toward the end of the conversation, Furrier inquired if there had been instances where Coursera engineered a solution that a form of was later rolled out by Amazon. Saeta commented on their video encoding and transcoding, even though Coursera opted for a 3^rd party provider even after Amazon’s Elastic transcoder was released. “But there are things that Amazon rolled out that we saw and thought, ‘Oh, if only they had done that sooner, our lives would have been so much easier.’”