This founder sold his last startup for $200 million - now he thinks Amazon and Google are 'weak'

Back in 2011, Citrix paid over $200 million for a startup called Cloud.com, a three-year-old startup that made software to turn any data center into a cloud, just like the kind Amazon Web Services has built into a $6 billion business.

Now, two of Cloud.com's founders - CEO Sheng Liang and Director of Engineering Will Chan - have a new startup, Rancher Labs, which recently raised a $10 million Series A round led by the Mayfield Fund.

And Liang thinks that big cloud vendors like Amazon and Google are "weak" and already way behind the curve.

"They're irrelevant because they're last generation," Liang says.

Rancher's technology is for management of software containers - a hot new technology led by $1 billion startup Docker, that lets developers take their code, wrap it up, and have it run on their laptop the same way it runs in a big data center or public cloud.

It's simple, but powerful, and developers love Docker, since it lets them focus on code and not worry about the servers it'll run on. It's a stark contrast from the Rancher founders' experience in the cloud market, where the choice of cloud computing vendors was forced on developers by operations or a business case, Liang says.

"For the first time, I've seen a piece of technology people voluntarily use," Liang says about Docker.

The whole key to the Docker phenomenon is portability: If it runs on my laptop, it'll run in the Amazon Web Services public cloud, where a swipe of the credit card gets you access to basically unlimited resources, without crashing or introducing new showstopping bugs.

There's an interesting side effect here: It doesn't really matter which vendor you go with. It'll run just as well as the Amazon Web Services cloud as it will on the Google Cloud Platform, which means developers will have no incentive to use one over the other.

It's not quite that easy, because lots of developers use vendor-specific cloud services like Google BigQuery or Amazon Elastic Beanstalk that lock people in and make it difficult to switch.

But Rancher Labs makes the "cloud overlay" software that helps developers make those moves, wrapping up additional services like databases and networking and making the entire application "as portable as Docker," Liang says.

Suddenly, developers would have their choice of cloud providers - choosing based on quality of service and reliability more than features or software support.

"I don't want to disrupt the incumbents," Liang says, "But I think abstracting the infrastructure is a good idea."

The only issue, Liang says, is proving the need for Rancher's services. So much of it hinges on Docker continuing to grow, and he acknowledges that all of this hype could backfire on the growing Docker-based startup scene.

"The only sure thing that kills a startup is being too early to market," Liang says.